Pansharpening under thin cloudy conditions is a practically significant yet rarely addressed task, challenged by simultaneous spatial resolution degradation and cloud-induced spectral distortions. Existing methods often address cloud removal and pansharpening sequentially, leading to cumulative errors and suboptimal performance due to the lack of joint degradation modeling. To address these challenges, we propose a Unified Pansharpening Model with Thin Cloud Removal (Pan-TCR), an end-to-end framework that integrates physical priors. Motivated by theoretical analysis in the frequency domain, we design a frequency-decoupled restoration (FDR) block that disentangles the restoration of multispectral image (MSI) features into amplitude and phase components, each guided by complementary degradation-robust prompts: the near-infrared (NIR) band amplitude for cloud-resilient restoration, and the panchromatic (PAN) phase for high-resolution structural enhancement. To ensure coherence between the two components, we further introduce an interactive inter-frequency consistency (IFC) module, enabling cross-modal refinement that enforces consistency and robustness across frequency cues. Furthermore, we introduce the first real-world thin-cloud contaminated pansharpening dataset (PanTCR-GF2), comprising paired clean and cloudy PAN-MSI images, to enable robust benchmarking under realistic conditions. Extensive experiments on real-world and synthetic datasets demonstrate the superiority and robustness of Pan-TCR, establishing a new benchmark for pansharpening under realistic atmospheric degradations.

Huachen Qi, Ruiyu Zhuo, Bowen Shi, Xiang Chang, Fei Chao, Changjing Shang, Qiang Shen. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Few-shot object detection (FSOD) poses a significant challenge due to the difficulty of learning robust and discriminative object representations under limited supervision. A widely adopted solution is the two-stage fine-tuning framework, wherein knowledge acquired from a large-scale base dataset is transferred to a novel dataset containing only a small number of labeled instances. However, this framework is prone to systematically misclassifying novel objects as background, primarily due to incorrect background label caused by the domain gap between base and novel datasets-an issue exacerbated by the sparse representation of novel categories. In this work, we show that this inherent weakness can be exploited by explicitly redefining the category structure and transferring the representations learned during the base training stage. Building on this insight, we propose a simple yet effective framework grounded in the Product of Experts (PoE) formulation, which estimates the joint distribution over background and novel categories by combining the unnormalized logits from independently trained classifiers. Notably, it does not require modifications of the base model or repetition of the base training phase. Furthermore, we introduce a strategy for identifying additional novel-category instances within the base dataset, which effectively augmenting the training set for fine-tuning. The resulting method is architecture-agnostic, imposes negligible overhead, and integrates seamlessly with existing two-stage fine-tuning pipelines. Extensive experiments on PASCAL VOC and COCO demonstrate that the proposed method yields consistent improvements across different baselines, achieving significant gains over state-of-the-art FSOD approaches.

Fuzzy rule interpolation (FRI) supports inference with sparse rule bases and plays a key role in Takagi–Sugeno–Kang (TSK) fuzzy model-based reasoning such as regression and real-time prediction. Current TSK-based FRI methods mainly employ distance-driven or rule-level optimisation strategies, implicitly assuming an unchanged geometric structure of the rule base. Recent findings, however, show that uneven variance allocation and directional imbalance in input data can induce rule selection bias, materially affecting interpolation behaviour. This paper introduces a Variance-Redistribution-Driven mechanism for TSK-based FRI algorithms. With a one-to-one orthogonal transformation in the input data supported by principal component analysis, the method redistributes variance across data dimensions while strictly preserving the original geometric and topological relationships of the data. The transformed variance structure propagates to the generated fuzzy rule base, yielding more directionally balanced rule selection and improved interpolation stability. The VRF mechanism functions as a preprocessing step and can be directly coupled with existing TSK-based FRI algorithms without increasing online computational complexity. This work not only delivers a simple and general optimisation method for a class of TSK-based FRI methods, but also extends the foundational theory of direction-aware FRI optimisation by clarifying how input-space variance reshaping influences rule induction and interpolation bias. Experiments on benchmark datasets confirm that VRF reliably enhances accuracy and robustness, while also exposing directional-sensitivity-dependent limitations that motivate further parameter-specific analysis for future.

Whilst Convolutional Neural Network (CNN)-based object tracking methods can achieve promising results on traditional well-lit datasets, it is challenging to accurately locate targets in low-light images taken in nighttime scenes, even for state-of-the-art (SOTA) trackers. Existing solutions often disregard potential image features beneficial for object tracking or focus solely on improving human perception, making it difficult to balance image enhancement and object tracking tasks. To address this issue and attain reliable nighttime unmanned aerial vehicle (UAV) tracking, we propose a lightweight Pyramid Attention-based low-light image enhancer, which serve as a plug-and-play solution before the trackers. In addition, we introduce a Pyramid Attention Module (PAM) to enhance the capability for multi-scale feature representation of images as image features are difficult to distinguish under low-light conditions. Experimental results reflect the effectiveness of our method in dealing with poor illumination situations.

Diffusion models have surpassed Generative Adversarial Networks in generating high-quality, high-resolution images, enhancing detail and diversity. However, diffusion models still demand significant time and computational resources. Current work indicates that the quality of generated images is tied to sampling time steps, with each phase in the generation process affecting training and output differently. To address this challenge, this study introduces evolutionary algorithms to optimize the sequence of time-step sampling probabilities within the training phase of diffusion models. Due to traditional sampling probability sequences involving floating points and high dimensions, this paper simplifies the search space and redefines the search objectives of the evolutionary algorithm, making the search process more efficient. Experimental results demonstrate that the proposed method not only speeds up the training process of diffusion models but also reveals that effective time step sampling probability sequences from adjacent training phases have similar distributions, indicating that a stable sequence of time steps exists that can consistently accelerate network convergence throughout extensive training phases. These findings not only enhance training efficiency but also reduce computational costs while maintaining the quality of generated images. The source code is available in the GitHub repository: https://github.com/zhangbeibei00/O2TDM.

Breast mass detection remains a significant challenge in developing effective computer-aided diagnosis (CADx) systems to assist clinicians in differentiating between benign and malignant masses. This paper introduces a novel fuzzy rule-based CADx approach for mammographic mass classification, utilising Transformation-based Fuzzy Rule Interpolation with Mahalanobis matrices (MT-FRI). This method enables reliable and interpretable classification by transforming attributes into a new feature space and interpolating for unmatched cases, making it well-suited to limited-data scenarios. The proposed approach integrates a structured pipeline encompassing feature extraction, feature selection, fuzzy rule generation, and interpolation inference, all designed to enhance transparency in diagnostic decisions. The system implementing the approach is evaluated on four widely-used mammographic datasets—INbreast, CBIS-DDSM, BCDR-D01, and BCDR-F01. For the first time, comparative experiments demonstrate that state-of-the-art fuzzy rule interpolative methods, particularly MT-FRI, achieve superior classification performance over representative classical machine learning models and deep neural networks. Unlike deep learning models, which require extensive labelled data and function as ”black boxes”, MT-FRI produces transparent, human-readable rules, supporting clinical interpretability. This work underscores the potential of MT-FRI as an adaptable and interpretable CADx solution for mammographic diagnosis, especially valuable in sparse-data environments.

Landscape painting is a gem of cultural and artistic heritage that showcases the splendor of nature through the deep observations and imaginations of its painters. Limited by traditional techniques, these artworks were confined to static imagery in ancient times, leaving the dynamism of landscapes and the subtleties of artistic sentiment to the viewer's imagination. Recently, emerging text-to-video (T2V) diffusion methods have shown significant promise in video generation, providing hope for the creation of dynamic landscape paintings. However, current T2V methods focus on generating natural videos, emphasizing the capture of details and the authenticity of physical laws. In contrast, landscape painting videos emphasize the overall dynamic aesthetic. Besides, challenges, such as the lack of specific datasets, the intricacy of artistic styles, and the creation of extensive, high-quality videos pose difficulties for these models in generating landscape painting videos. In this article, we propose landscape painting videos-high definition (LPV-HD), a novel T2V dataset for landscape painting videos, and noise-aware diffusion model (NADM), a T2V model that utilizes Stable Diffusion. Specifically, we present a motion module featuring a dual attention mechanism to capture the dynamic transformations of landscape imageries, alongside a noise adapter to leverage unsupervised contrastive learning in the latent space to ensure the overall beauty of the landscape painting video. Following the generation of keyframes, we employ optical flow for frame interpolation to enhance video smoothness. Our method not only retains the essence of the landscape painting imageries but also achieves dynamic transitions, significantly advancing the field of artistic video generation. Source code and dataset are available at https://github.com/llzlh21/NADM.

Fuzzy rule interpolation (FRI) has been successfully applied to address real-world problems where domain knowledge is incomplete, particularly in situations where observations do not directly match existing rules. However, most of the existing research focuses on fuzzy systems based on Mamdani models. Recently, a pioneering approach has been proposed for rule interpolation within Takagi–Sugeno–Kang (TSK) fuzzy models, employing two distinct techniques that cluster neighboring rules to facilitate interpolated outcomes. Although promising, these techniques have been developed through empirical methods, lacking formal representation and theoretical justification. This article formalizes such emerging empirical studies with the aim of strengthening the theoretical foundation of FRI. It introduces the location view of rules and rule bases regarding TSK models, analyzing the influence of individual rules within FRI techniques for such models. Through geometrizing sparse rule bases, comparative investigations of the empirical and theoretical results are carried out. Moreover, geometric projection of the rules is combined with a rule weight adjustment mechanism to reinforce the capability of the formalized FRI techniques. By embedding directional parameters into the rules, the resulting method increases the weighting of critical rules, strengthening the FRI algorithms' performance when many rules may be selected for interpolation. Experimental results confirm the effectiveness of the theoretical model and the consistency with the existing empirical findings. This highlights the potential and significance of formal studies in guiding algorithmic improvements on FRI with TSK models.

Few-shot object detection (FSOD) within remote sensing imagery has achieved great advancements in recent years. However, most existing methods are facing one key challenge while handling remote sensing images: many unlabeled instances in few-shot images are treated as background, which tends to degrade the generalization of the trained model severely. This paper presents HPMF, a Hypergraph-guided Prototype Mining Framework that addresses the challenge through joint optimization from three perspectives. The first is Hierarchical Reference Mining (HRM) which constructs a class-instance dual-driven prototype space that enables mining the unlabeled instances via cross-hierarchical similarity fusion. The second is a Robust Pseudo-box Estimator (RPE) that generates high-quality pseudo bounding boxes for the HRM-mined instances via adaptive density clustering and multi-statistic aggregation. The third is a Hypergraph-Guided Decoder (HGD) that introduces hypergraphs into the transformer decoder for group semantic modeling, enhancing high-order semantic association and similarity of instance features, thereby further improving the mining performance of the HRM module. Extensive experiments under various settings show that the proposed HPMF outperforms state-of-the-art methods consistently across multiple widely adopted remote-sensing FSOD benchmarks such as DIOR, NWPU-VHR10 v2, and HRRSD.

Nonlinear systems, such as robotic systems, play an increasingly important role in our modern daily life and have become more dominant in many industries; however, robotic control still faces various challenges due to diverse and unstructured work environments. This article proposes a double-loop recurrent neural network (DLRNN) with the support of a Type-2 fuzzy system and a self-organizing mechanism for improved performance in nonlinear dynamic robot control. The proposed network has a double-loop recurrent structure, which enables better dynamic mapping. In addition, the network combines a Type-2 fuzzy system with a double-loop recurrent structure to improve the ability to deal with uncertain environments. To achieve an efficient system response, a self-organizing mechanism is proposed to adaptively adjust the number of layers in a DLRNN. This work integrates the proposed network into a conventional sliding mode control (SMC) system to theoretically and empirically prove its stability. The proposed system is applied to a three-joint robot manipulator, leading to a comparative study that considers several existing control approaches. The experimental results confirm the superiority of the proposed system and its effectiveness and robustness in response to various external system disturbances.

Abstract Image Super-Resolution (ISR) is utilised to generate a high-resolution image from a low-resolution one. However, most current techniques for ISR confront three main constraints: i) the assumption that there is sufficient data available for training, ii) the presumption that areas of the images concerned do not involve missing data, and iii) the development of a computationally efficient model that does not compromise performance. In addressing these issues, this study proposes a novel lightweight approach termed Fuzzy Rough Feature Selection-based ANFIS Interpolation (FRFS-ANFISI) for ISR. Popular feature extraction algorithms are employed to extract the potentially significant features from images, and population-based search mechanisms are utilised to implement effective FRFS methods that assist in selecting the most important features among them. Subsequently, the processed data is entered into the ANFIS interpolation model to execute the ISR operation. To tackle the sparse data challenge, two adjacent ANFIS models are trained with sufficient data where appropriate, intending to position the ANFIS model of sparse data in the middle. This enables the two neighbouring ANFIS models to be interpolated to produce the otherwise missing knowledge or rules for the model in between, thereby estimating the corresponding outcomes. Conducted on standard ISR benchmark datasets while considering both sufficient and sparse data scenarios, the experimental studies demonstrate the efficacy of the proposed approach in helping deal with the aforementioned challenges facing ISR.

Abstract Representation‐based classification (RC) is an effective gauge of data similarity between a single instance and the whole dataset, which extends traditional individual‐wise distance metrics using representation coefficients. These coefficients show remarkable discrimination nature via various regularisation terms, but the interference from potentially uncorrelated objects involved in this single‐to‐global relation can degrade the effectiveness of the coefficients. In order to filter out those unproductive, or even counter‐productive, information from the decision making processes, this paper proposes a local representation‐based classification (LRC) algorithm to improve the classification accuracy or the RC approach. LRC uses a single‐to‐local relation induced by the local representation‐based neighbourhood (LRN) of each object, rather than the single‐to‐global relationship used by RC. Thanks to LRN, a compact and relevant dataset can be formed by selecting the most relevant data instances in the original dataset, to render a robust representation of a query. LRC was applied to multiple publicly available datasets, and the experimental results demonstrate the superiority of the proposed LRC algorithm as evidenced by the higher classification accuracy and more noise‐tolerant capability in reference to alternative RC approaches. Moreover, the sampling ability of LRN is also verified via a comparative study.

Approximation biases of value functions are considered a key problem in reinforcement learning (RL). In particular, existing RL algorithms are hindered by overestimation and underestimation biases, i.e., value mismatching between RL's actual returns and action-value approximations limits the performance of RL algorithms. In this article, we first develop a new synthesis loss function for RL's action-value estimation integrating a regularization term and a modified "clipped double Q-learning" structure for solving overestimation and underestimation biases. To minimize the differences between action-value estimations and actual returns in RL, we develop a new discrepancy function to determine the type and magnitude of approximation biases. Then, two coefficients embedded in the synthesis loss are automatically tuned by minimizing the discrepancy function during training to minimize approximation biases. We further design a new actor-critic (AC) algorithm, named AC with synthesis loss (ACSL), by integrating the synthesis loss function and an error-controlled mechanism. Experimental results on continuous control tasks illustrate that the proposed ACSL algorithm outperforms other cutting-edge RL methods in many tasks and that the proposed synthesis loss function is easily implemented into other algorithms and significantly reduces approximation biases while improving performance. The proposed method can successfully handle many complex continuous control tasks and can greatly outperform other state-of-the-art algorithms on most tasks.

The writing sequence of numerals or letters often affects aesthetic aspects of the writing outcomes. As such, it remains a challenge for robotic calligraphy systems to perform, mimicking human writers' implicit intention. This article presents a new robot calligraphy system that is able to learn writing sequences with limited sequential information, producing writing results compatible to human writers with good diversity. In particular, the system innovatively applies a gated recurrent unit (GRU) network to generate robotic writing actions with the support of a prelabeled trajectory sequence vector. Also, a new evaluation method is proposed that considers the shape, trajectory sequence, and structural information of the writing outcome, thereby helping ensure the writing quality. A swarm optimization algorithm is exploited to create an optimal set of parameters of the proposed system. The proposed approach is evaluated using Arabic numerals, and the experimental results demonstrate the competitive writing performance of the system against state-of-the-art approaches regarding multiple criteria (including FID, MAE, PSNR, SSIM, and PerLoss), as well as diversity performance concerning variance and entropy. Importantly, the proposed GRU-based robotic motion planning system, supported with swarm optimization can learn from a small dataset, while producing calligraphy writing with diverse and aesthetically pleasing outcomes.

Fuzzy rule interpolation (FRI) empowers fuzzy rule-based systems (FRBSs) with the ability to infer, even when presented with a sparse rule base where no direct rules are applicable to a given observation. The core principle lies in creating an intermediate fuzzy rule-either interpolated or extrapolated-derived from rules neighboring the observation. Conventionally, the selection of these rules hinges upon distance metrics. While this approach is easy to grasp and has been instrumental in the evolution of various FRI methods, it is burdened by the necessity of extensive distance calculations. This becomes particularly cumbersome when swift responses are imperative or when dealing with large datasets. This article introduces a groundbreaking rule-ranking-based FRI method, termed RT-FRI, which overcomes the constraints of the longstanding distance-centric FRI approach. Instead of relying on distances, RT-FRI harnesses ranking scores for rules and unmatched observation. These scores are produced by amalgamating the antecedent attributes using aggregation functions, thus streamlining the rule selection procedure. Recognizing the rigid monotonicity demands of aggregation functions, a variant-DMRT-FRI-has been introduced to ensure directional monotonicity. Experimental results indicate that RT-FRI emerges as a highly efficient technique, with DMRT-FRI exemplifying a notable balance of accuracy and efficiency.

Learning calligraphy writing skills in robots is regarded as a sophisticated task. Current robotic researchers have proposed many methods to implement various robotic calligraphy systems. However, several limitations of these methods, such as high computational costs and few diversities of generated results constrain the development of calligraphy robots. This article proposes a robotic writing framework based on a robotic hand–eye coordination method to solve these limitations. Inspired by the internal model control (IMC) system, a vision-motor network and a motor-vision network are built to simulate the direct and reverse models, respectively, in the IMC system of a robotic manipulator. The vision-motor network works as an action generator to convert image inputs to robotic actions, and the motor-vision network assists in the training of the vision-motor network. Thus, a pretraining of the motor-vision network is established by random writing movements of a robotic manipulator. Experimental results demonstrate that the proposed method can successfully write strokes of Chinese characters by inputting target stroke images. Although the proposed method is applied to robotic calligraphy, the underpinning research is readily applicable to many other applications, such as human–robot motion mimicking.

Despite the rapid advance in multispectral (MS) pansharpening, existing convolutional neural network (CNN)-based methods require training on separate CNNs for different satellite datasets. However, such a single-task learning (STL) paradigm often leads to overlooking any underlying correlations between datasets. Aiming at this challenging problem, a multitask network (MTNet) is presented to accomplish joint MS pansharpening in a unified framework for images acquired by different satellites. Particularly, the pansharpening process of each satellite is treated as a specific task, while MTNet simultaneously learns from all data obtained from these satellites following the multitask learning (MTL) paradigm. MTNet shares the generic knowledge between datasets via task-agnostic subnetwork (TASNet), utilizing task-specific subnetworks (TSSNets) to facilitate the adaptation of such knowledge to a certain satellite. To tackle the limitation of the local connectivity property of the CNN, TASNet incorporates Transformer modules to derive global information. In addition, band-aware dynamic convolutions (BDConvs) are proposed that can accommodate various ground scenes and bands by adjusting their respective receptive field (RF) size. Systematic experimental results over different datasets demonstrate that the proposed approach outperforms the existing state-of-the-art (SOTA) techniques.

Deep convolutional neural networks have delivered remarkable aptitude in performing Synthetic Aperture Radar (SAR) image speckle removal tasks. Such approaches are nevertheless constrained in balancing speckle removal and preservation of spatial information, particularly with respect to strong speckle noise. In this paper, a novel residual-in-residual dense generative adversarial network is proposed to effectively suppress SAR image speckle while retaining rich spatial information. A despeckling sub-network composed of residual-in-residual dense blocks with an encoder-decoder structure is devised to learn end-to-end mapping of noisy images onto noise-free images, where the combination of residual-in-residual structure and dense connection significantly enhances the feature representation capability. In addition, a discriminator sub-network with a fully convolutional structure is introduced, and the adversarial learning strategy is adopted to continuously refine the quality of despeckled results. Systematic experimental results on simulated and real SAR images demonstrate that the novel approach offers superior performance in both quantitative and visual evaluation as compared to state-of-the-art methods.

Despite recent advances in video-guided robotic imitation learning, many methods still rely on human experts to provide sparse rewards that indicate whether robots have successfully completed tasks. The challenge of enabling robots to autonomously evaluate whether their actions can complete complex, multi-stage tasks remains unresolved. In this work, we propose an efficient few-shot robotic learning algorithm that centres around learning and evaluating from a third-person perspective to address the aforementioned challenge. We develop a novel Siamese neural network-based robotic action-state evaluation system, named “Behavior-Outcome Dual Assessment” (BODA), in our robotic imitation learning system, so as to replace artificial evaluations from human experts in multi-stage imitation learning processes and to improve learning efficiency. In this way, one video demonstration of a target task is divided into several stages. For each stage, we design two Siamese neural network-based evaluation modules in BODA: One module focuses on action changes, and the other handles working environment changes. The two modules work together to provide a comprehensive assessment of the robot's completion of each stage from the view of both the action and working environment changes. Then, BODA is integrated within a model-based reinforcement learning framework to enable the completion of our imitation learning cycle. Extensive experiments demonstrate that the evaluation processes of BODA can automatically and accurately evaluate task completion status without human intervention. In contrast to conventional methods, BODA is able to keep the accumulation of errors within acceptable limits through self-assessment in stages.

Abstract Fuzzy rule interpolation (FRI) offers a reliable approach for providing an interpretable approximate decision with a sparse rule base, when a new observation does not match any existing rules. As the mainstream application of a fuzzy rule base is to extract valuable approximate information from each individual rules, existing FRI methods typically work by postulating that the more rules used to implement the interpolation the better the reasoning outcomes. Yet, empirical results have shown that using a large number of rules in an FRI process may adversely lead to worsening the accuracy of the inference outcomes, not just degrading efficiency. The objective of this work is to set a firm theoretical foundation for the eventual establishment of a novel FRI approach. It achieves this goal by mapping the structural patterns within a given fuzzy rule base onto a mathematically isomorphic data space, such that the essential information embedded in the original rule base can be effectively captured, represented and analysed. The resulting mathematically mapped patterns enable the production of a theorem that determines the upper limit of the number of rules required to effectively and efficiently perform FRI. The experimental investigations reported herein demonstrate that the number of required rules to perform FRI obeys the theorem discovered in this work.

Deep learning reveals excellent potential for accomplishing change detection in SAR imagery. Yet, it suffers from the problem of requiring large amounts of labeled samples, whilst labeling SAR imagery for change detection requires experts to label individual images at the pixel level , which is extremely tedious and time-consuming. Also, sample imbalance continues to present a serious challenge for the existing change detection techniques. To tackle these problems, in this study, a Deep Collaborative semi-supervised learning Framework with Class-Rebalancing (DCF-CRe) is proposed for SAR imagery change detection, by exploiting Convolutional Neural Network (CNN) and deep clustering. In particular, a Siamese Difference Fusion Network (SDFNet) is devised to implement change detection while effectively reducing the information loss due to the generation of difference images and highlighting features of the changed regions.In so doing, only a tiny batch of labeled samples is utilized to train SDFNet in order to obtain predicted change map and deep features. In addition, the Approximate Rank-Order Clustering (AROC) algorithm is employed to cluster the deep features, generating pseudo-labels for abundant unlabeled samples . DCF-CRe is then applied to select appropriate pseudo-labels and to add labeled samples to train SDFNet. Experimental results evaluated on six challenging datasets show that this proposed approach can achieve performance superior to state-of-the-art change detection methods for SAR imagery.

Abstract Image processing is a very broad field containing various areas, including image super-resolution (ISR) which re-represents a low-resolution image as a high-resolution one through a certain means of image transformation. The problem with most of the existing ISR methods is that they are devised for the condition in which sufficient training data is expected to be available. This article proposes a new approach for sparse data-based (rather than sufficient training data-based) ISR, by the use of an ANFIS (Adaptive Network-based Fuzzy Inference System) interpolation technique. Particularly, a set of given image training data is split into various subsets of sufficient and sparse training data subsets. Typical ANFIS training process is applied for those subsets involving sufficient data, and ANFIS interpolation is employed for the rest that contains sparse data only. Inadequate work is available in the current literature for the sparse data-based ISR. Consequently, the implementations of the proposed sparse data-based approach, for both training and testing processes, are compared with the state-of-the-art sufficient data-based ISR methods. This is of course very challenging, but the results of experimental evaluation demonstrate positively about the efficacy of the work presented herein.

Metalearning has been widely applied for implementing few-shot learning and fast model adaptation. Particularly, existing metalearning methods have been exploited to learn the control mechanism for gradient descent processes, in an effort to facilitate gradient-based learning in gaining high speed and generalization ability. This article presents a novel method that controls the gradient descent process of the model parameters in a neural network, by limiting the model parameters within a low-dimensional latent space. The main challenge for implementing this idea is that a decoder with many parameters may be required. To tackle this problem, the article provides an alternative design of the decoder with a structure that shares certain weights, thereby reducing the number of required parameters. In addition, this work combines ensemble learning with the proposed approach to improve the overall learning performance. Systematic experimental studies demonstrate that the proposed approach offers results superior to the state of the art in performing the Omniglot classification and miniImageNet classification tasks.

Dealing with uncertain rail disruptions effectively raises a significant challenge for computational intelligence research. This article studies the bus bridging problem under demand uncertainty, where the passenger demand is represented as parametric interval-valued fuzzy variables and their associated uncertainty distribution sets. A distributionally robust fuzzy optimization model is proposed to minimize the maximum travel time and to search for the optimal scheme for vehicle allocation, route selection, and frequency determination. To solve the proposed robust model, we discuss the computational issues concerning credibilistic constraints, turning the robust counterpart model into computationally tractable equivalent formulations. The proposed approach is verified, and the resulting method is validated with a report on uncertain parameters in a real-world disrupted event of Shanghai Rail Line 1. Experimental results show that the distributionally robust fuzzy optimization approach can provide a better uncertainty-immunized solution.

Fuzzy clustering decomposes data into clusters using partial memberships by exploring the cluster structure information, which demonstrates the comparable performance for knowledge exploitation under the circumstance of information incompleteness. In general, this scheme considers the memberships of objects to cluster centroids and applies to clusters with the spherical distribution. In addition, the noises and outliers may significantly influence the clustering process; a common mitigation measure is the application of separate noise processing algorithms, but this usually introduces multiple parameters, which are challenging to be determined for different data types. This article proposes a new fuzzy-rough intrigued harmonic discrepancy clustering (HDC) algorithm by noting that fuzzy-rough sets offer a higher degree of uncertainty modeling for both vagueness and imprecision present in real-valued datasets. The HDC is implemented by introducing a novel concept of harmonic discrepancy, which effectively indicates the dissimilarity between a data instance and foreign clusters with their distributions fully considered. The proposed HDC is thus featured by a powerful processing ability on complex data distribution leading to enhanced clustering performance, particularly on noisy datasets, without the use of explicit noise handling parameters. The experimental results confirm the effectiveness of the proposed HDC, which generally outperforms the popular representative clustering algorithms on both synthetic and benchmark datasets, demonstrating the superiority of the proposed algorithm.

Fuzzy rule interpolation (FRI) strongly supports approximate inference when a new observation matches no rules, through selecting and subsequently interpolating appropriate rules close to the observation from the given (sparse) rule base. Traditional ways of implementing the critical rule selection process are typically based on the exploitation of Euclidean distances between the observation and rules. It is conceptually straightforward for implementation but applying this distance metric may systematically lead to inferior results because it fails to reflect the variations of the relevance or significance levels among different domain features. To address this important issue, a novel transformation-based FRI approach is presented, on the basis of utilizing the Mahalanobis distance metric. The new FRI method works by transforming a given sparse rule base into a coordinates system where the distance between instances of the same category becomes closer while that between different categories becomes further apart. In so doing, when an observation is present that matches no rules, the most relevant neighboring rules to implement the required interpolation are more likely to be selected. Following this, the scale and move factors within the classical transformation-based FRI procedure are also modified by Choquet integral. Systematic experimental investigation over a range of classification problems demonstrates that the proposed approach remarkably outperforms the existing state-of-the-art FRI methods in both accuracy and efficiency.

Fuzzy Rule Interpolation (FRI) provides a useful mechanism to derive reasonable approximate inference outcomes when presented with a sparse rule base. However, conventional FRI techniques are exploited to generate fuzzy conclusions for observations that have no matching rules in a static rule base. Any interpolated rules created during the interpolation process are abandoned once the interpolated results are obtained, despite they may contain valuable, and general or generalisable, information about the domain problem. This paper proposes an approach that helps improve the sparse rule base dynamically, through measuring and subsequently adding certain interpolated rules that are deemed of high value into the sparse rule base. In particular, the value of an interpolated rule is assessed via its similarities with those given in the sparse rule base, in an effort to determine whether it is to be incorporated in the existing rule base. Experimental results on benchmark datasets illustrate that by the use of a dynamically enriched rule base the popular transformation-based FRI algorithm is able to produce strengthened inference outcomes over those that it generates while using just the original static rule base.

Oracle Bone Inscription (OBI) is an early hieroglyph in China, which is the most famous ancient writing system in the world. However, only a small number of OBI characters have been fully deciphered today. Chinese characters have different forms in different historical stages; therefore, it is very difficult to directly translate OBI characters to modern Chinese characters due to the long historic evolutionary process. In this paper, we propose a cascade generative adversarial networks (GAN) framework for deciphering OBI characters, named "Sundial-GAN'', which is a cascaded structure to simulate Chinese characters' evolutionary process from an OBI character to its potential modern Chinese character. We select four representative stages in the evolutionary process of OBI, each of which is implemented by an individual GAN structure based on the characteristics of each evolutionary stage. These structures are cascaded in sequence to accurately simulate the Chinese characters' evolutionary process. For each input OBI character, Sundial-GAN can successfully generate the input's different forms at the four historical stages. Extensive experiments and comparisons demonstrate that generated characters at each stage have high similarities with real existing characters; therefore, the proposed method can significantly improve the efficiency and accuracy of OBI deciphering for archaeological researchers. Compared to direct image-to-image translation methods, our approach allows for a smoother translation process, a better grasp of details, and more effective avoiding random mappings in GANs.

Computer aided diagnosis (CAD) systems play an essential role in the early detection and diagnosis of developing disease for medical applications. In order to obtain the highly recognizable representation for the medical images, a self-adaptive discriminative autoencoder (SADAE) is proposed in this paper. The proposed SADAE system is implemented under a deep metric learning framework which consists of$K$local autoencoders, employed to learn the$K$subspaces that represent the diverse distribution of the underlying data, and a global autoencoder to restrict the spatial scale of the learned representation of images. Such community of autoencoders is aided by a self-adaptive metric learning method that extracts the discriminative features to recognize the different categories in the given images. The quality of the extracted features by SADAE is compared against that of those extracted by other state-of-the-art deep learning and metric learning methods on five popular medical image data sets. The experimental results demonstrate that the medical image recognition results gained by SADAE are much improved over those by the alternatives.

Numerous multiobjective evolutionary algorithms (EAs) have been designed for constrained optimization over the past two decades. The idea behind these algorithms is to transform constrained optimization problems (COPs) into multiobjective optimization problems without any constraint, and then solve them. In this article, we propose a new multiobjective method for constrained optimization, which works by converting a COP into a problem with helper and equivalent objectives. An equivalent objective means that its optimal solution set is the same as that of the constrained problem but a helper objective does not. Then, this multiobjective optimization problem is decomposed into a group of subproblems using the weighted sum approach. Weights are dynamically adjusted so that each subproblem eventually tends to a problem with an equivalent objective. We theoretically analyze the computational time of the helper and equivalent objective method on a hard problem called "wide gap." In a wide gap problem, an algorithm needs exponential time to cross between two fitness levels (a wide gap). We prove that using helper and equivalent objectives can shorten the time of crossing the wide gap. We conduct a case study for validating our method. An algorithm with helper and equivalent objectives is implemented. The experimental results show that its overall performance is ranked first when compared with other eight state-of-the-art EAs on IEEE CEC2017 benchmarks in constrained optimization.

Discriminative correlation filter (DCF) has contributed tremendously to address the problem of object tracking benefitting from its high computational efficiency. However, it has suffered from performance degradation in unmanned aerial vehicle (UAV) tracking. This article presents a novel semantic-aware real-time correlation tracking framework (SARCT) for UAV videos to enhance the performance of DCF trackers without incurring excessive computing cost. Specifically, SARCT first constructs an additional detection module to generate ROI proposals and to filter any response regarding the target irrelevant area. Then, a novel semantic segmentation module based on semantic template generation and semantic coefficient prediction is further introduced to capture semantic information, which can provide precise ROI mask, thereby effectively suppressing background interference in the ROI proposals. By sharing features and specific network layers for object detection and semantic segmentation, SARCT reduces parameter redundancy to attain sufficient speed for real-time applications. Systematic experiments are conducted on three typical aerial datasets in order to evaluate the performance of the proposed SARCT. The results demonstrate that SARCT is able to improve the accuracy of conventional DCF-based trackers significantly, outperforming state-of-the-art deep trackers.

An adaptive-network-based fuzzy inference system (ANFIS) offers a popular and powerful fuzzy inference mechanism. As with many other advanced data-driven techniques, developing an effective ANFIS typically requires sufficient training data. However, in many real-world applications, it is not always straightforward to obtain a large amount of representative data that cover the entire problem space to accomplish the required training, seriously restricting the performance of a learned ANFIS. This article introduces a new ANFIS learning approach through an evolutionary process, which is able to generate an ANFIS with only a small amount of training data in a certain problem region, by interpolating well-trained ANFISs in the neighboring regions. Such a process works by first producing an initial population of candidate fuzzy rules in the region of data shortage, through interpolating a rule dictionary constructed from trained ANFISs in the neighborhood regions. The crossover and mutation operations over these candidate rules are then executed in an effort to attain candidates of improved performance. When this genetic learning process terminates, the chromosomes in the final population either collectively form or each individually represents a learned ANFIS, depending on whether a single fuzzy rule or a set of fuzzy rules representing an entire ANFIS is implemented with a chromosome within the evolving population. Comparative experimental evaluations on both synthetic and real-world datasets are carried out, demonstrating that in spite of data shortage, the proposed interpolation approach is able to produce ANFIS models that significantly outperform those trained using existing learning mechanisms.

When a fuzzy system is presented with an incomplete (or sparse) rule base, fuzzy rule interpolation (FRI) offers a useful mechanism to infer conclusions for unmatched observations. However, most existing FRI methodologies are established for Mamdani inference models, but not for Takagi–Sugeno–Kang (TSK) ones. This article presents a novel approach for computing interpolated outcomes with TSK models, using only a small number of neighboring rules to an unmatched observation. Compared with existing methods, the new approach helps improve the computational efficiency of the overall interpolative reasoning process, while minimizing the adverse impact on accuracy induced by firing those rules of low similarities with the new observation. For problems that involve a rule base of a large size, where closest neighboring rules may be rather alike to one another, a rule-clustering-based method is introduced. It derives an interpolated conclusion by first clustering rules into different groups with a clustering algorithm and then, by utilizing only those rules that are each selected from one of a given, small number of closest rule clusters. Systematic experimental examinations are carried out to verify the efficacy of the introduced techniques, in comparison with state-of-the-art methods, over a range of benchmark regression problems, while employing different clustering algorithms (which also shows the flexibility in ways of implementing the novel approach).

Multispectral (MS) pan-sharpening aims at producing high-resolution (HR) MS images in both spatial and spectral domains, by merging single-band panchromatic (PAN) images and corresponding MS images with low spatial resolution. The intuitive way to accomplish such MS pan-sharpening tasks, or to reconstruct ideal HR-MS images, is to extract feature pairs from the given PAN and MS images and to fuse the results. Therefore, feature extraction and feature fusion are two key components for MS pan-sharpening. This article presents a novel MS pan-sharpening network (MPNet), including a heterogeneous pair of feature extraction pathways (FEPs) and a convolutional long short-term memory (ConvLSTM)-based hierarchical feature fusion module (HFFM). Specifically, we design a PAN FEP to extract 2-D feature maps via 2-D convolutions and dual attention, while an MS FEP is introduced in an effort to obtain 3-D representations of MS image by 3-D convolutions and triple attention. To merge the resulting hierarchical features, the ConvLSTM-based HFFM is developed, leveraging intralevel fusion, interlevel fusion, and information exchange within one single framework. Here, the interlevel fusion is implemented with the ConvLSTM to capture the dependencies among hierarchical features, reduce redundant information, and effectively integrate them via its recurrent architecture. The information exchange between different FEPs helps enhance the representations for subsequent processing. Systematic comparative experiments have been conducted on three publicly available datasets at both reduced resolution and full resolution, demonstrating that the proposed MPNet outperforms state-of-the-art methods in the literature.

Inertial measurement units (IMUs) have been widely used to provide accurate location and movement measurement solutions, along with the advances of modern manufacturing technologies. The scale factors of accelerometers and gyroscopes are linear when the range of the sensors are reasonably small, but the factor becomes nonlinear when the range gets much bigger. Based on this observation, this article presents a calibration method for low-cost IMU by effectively deriving the nonlinear scale factors of the sensors. Two motion patterns of the sensor on a rigid object are moved to collect data for calibration: One motion pattern is to upcast and rotate the rigid object, and another pattern is to place the rigid object on a stable base in different attitudes. The rotation motion produces centripetal and Coriolis force, which increases the measurement range of accelerometers. Four cost functions with different weight factors and two sets of data are utilized to optimize the IMU parameters. The weight factor comes from derived formula with input values which are the variance of the noise of the sampled data. The proposed approach was validated and evaluated on both synthetic and real-world data sets, and the experimental results demonstrated the superiority of the proposed approach in improving the accuracy of IMU for long-range use. In particular, the errors of acceleration and angular velocity led by our algorithm are significantly smaller than those resulted from the existing approaches using the same testing data sets, demonstrating a remarkable improvement of 64.12% and 47.90%, respectively.

Traditional fuzzy rule interpolation (FRI) methods typically utilise Euclidean distances between an observation and the rules in a given sparse rule base to select a set of rules closest to the observation to perform interpolation. However, simply applying the Euclidean distance metric may frequently lead to inferior results, because it cannot take into consideration the relevance degree of different features. To address this important issue, this paper presents an initial framework for a novel FRI approach which works based on exploiting a transformed rule base. Mahalanobis matrix learned by metric learning methods is used herein to transform the given sparse rule base to a new coordinates system where the distance between instances of the same category is closer, and instances of different categories is distant from each other. When a new observation is present which matches no rules, the selection of the nearest rules to implement the required interpolation is carried out in the transformed coordinates system. Then, the scale and move factors within the classical transformation-based FRI procedure are modified by Choquet integral. Experimental results obtained by employing different metric learning methods and Choquet integral over seven classification problems demonstrate that the proposed approach remarkably outperforms existing FRI methods.

Recently proposed methods of TSK inference extension (TSK+) and K closest rules based TSK (KCR) are able to potentially perform fuzzy rule-based inference with a sparse rule base, for regression problems. However, in certain real-world applications, observations may be rather far away from any rules that may be fired in order to implement the required regression and hence, these state-of-the-art techniques may not work satisfactorily. To investigate an alternative to address such practical problems, this paper presents a novel fuzzy rule interpolation (FRI) approach. It works based on integrating feature selection and feature aggregation to attain a rule-ranking list within the given sparse rule base. The process of acquiring an ordered rule list is accomplished offline, with an ordered rule base ready for use prior to the arrival of any observation online. It boosts the computational efficiency of closest rule selection during the FRI process. Initial experimental results demonstrate that the proposed approach is able to address the problems that TSK+ and KCR fail to do while having a similar time efficiency to them. Further, only two nearest rules to the observation are required to derive interpolated results, thereby significantly improving the automation level of the interpolative reasoning system.

Apart from the need for superior accuracy, healthcare applications of intelligent systems also demand the deployment of interpretable machine learning models which allow clinicians to interrogate and validate extracted medical knowledge. Fuzzy rule-based models are generally considered interpretable that are able to reflect the associations between medical conditions and associated symptoms, through the use of linguistic if-then statements. Systems built on top of fuzzy sets are of particular appealing to medical applications since they enable the tolerance of vague and imprecise concepts that are often embedded in medical entities such as symptom description and test results. They facilitate an approximate reasoning framework which mimics human reasoning and supports the linguistic delivery of medical expertise often expressed in statements such as 'weight low' or 'glucose level high' while describing symptoms. This paper proposes an approach by performing data-driven learning of accurate and interpretable fuzzy rule bases for clinical decision support. The approach starts with the generation of a crisp rule base through a decision tree learning mechanism, capable of capturing simple rule structures. The crisp rule base is then transformed into a fuzzy rule base, which forms the input to the framework of adaptive network-based fuzzy inference system (ANFIS), thereby further optimising the parameters of both rule antecedents and consequents. Experimental studies on popular medical data benchmarks demonstrate that the proposed work is able to learn compact rule bases involving simple rule antecedents, with statistically better or comparable performance to those achieved by state-of-the-art fuzzy classifiers.

Abstract Dynamic fuzzy rule interpolation (D‐FRI) technique delivers a dynamic rule base through the utilisation of fuzzy rule interpolation to infer more accurate results for a given application problem. D‐FRI offered dynamic rule base is very useful in security areas where network conditions are always volatile and require the most updated rule base. A honeypot is a vital part of any security infrastructure for directly investigating attacks and attackers in real‐time to strengthen the overall security of the network. However, a honeypot as a concealed system can only function successfully while its identity is not revealed to any attackers. Attackers always attempt to uncover such honeypots for avoiding any trap and strengthening their attacks. Active fingerprinting attack is used to detect these honeypots by injecting purposefully designed traffic to a network. Such an attack can be prevented by controlling the traffic but this will make honeypot unusable system if its interaction with the outside world is limited. Alternatively, it is practically more useful if this fingerprinting attack is detected in real‐time to manage its immediate consequences and preventing the honeypot. This article offers an approach to building a cognizant honeypot for detecting active fingerprinting attacks through the utilisation of the established D‐FRI technique. It is based on the use of just a sparse rule base while remaining capable of detecting active fingerprinting attacks when the system does not find any matching rules. Also, it learns from current network conditions and offers a dynamic rule base to facilitate more accurate and efficient detection.

A major assumption for constructing an effective adaptive-network-based fuzzy inference system (ANFIS) is that sufficient training data are available. However, in many real-world applications, this assumption may not hold, thereby requiring alternative approaches. In light of this observation, this article focuses on automated construction of ANFISs in an effort to enhance the potential of the Takagi-Sugeno fuzzy regression models for situations where only limited training data are available. In particular, the proposed approach works by interpolating a group of fuzzy rules in a certain given domain with the assistance of existing ANFISs in its neighboring domains. The construction process involves a number of computational mechanisms, including a rule dictionary which is created by extracting the rules from the existing ANFISs; a group of rules which are interpolated by exploiting the local linear embedding algorithm to build an intermediate ANFIS; and a fine-tuning method which refines the resulting intermediate ANFIS. The experimental evaluation on both synthetic and real-world datasets is reported, demonstrating that when facing the data shortage situations, the proposed approach helps significantly improve the performance of the original ANFIS modeling mechanism.

Reasonable risk models in hazardous material transportation are of practical significance, for safeguarding the lives and properties, protecting the natural environment, and facilitating sustainable development. The existing risk models can be classified into summation risk and maximum risk models, which result in overreliance on overall or local risk. To overcome these problems, in this article, we present two novel risk models considering different aggregation methods on local risks. The first model is supported by an ordered weighted averaging (OWA) operator, which assigns the weights according to the position of the segment risk in the process of risk aggregation, and the second model is supported by state variable weight (SVW) vector, which adjusts the weights on segments according to the change of segment risk values. Generally speaking, an OWA risk model is used under the situation with complete weighting information, whereas an SVW risk model could be used under the situation with incomplete weighting information. Based on the analysis for variable weight mechanism, we show that both models could effectively balance the overall risk with the local risks assisted by weights variety. Numerical experiments are provided to illustrate the validity of the proposed risk models.

It has been a challenging task for a robotic arm to accurately reach and grasp objects, which has drawn much research attention. This article proposes a robotic hand-eye coordination system by simulating the human behavior pattern to achieve a fast and robust reaching ability. This is achieved by two neural-network-based controllers, including a rough reaching movement controller implemented by a pretrained radial basis function for rough reaching movements, and a correction movement controller built from a specifically designed brain emotional nesting network (BENN) for smooth correction movements. In particular, the proposed BENN is designed with high nonlinear mapping ability, with its adaptive laws derived from the Lyapunov stability theorem; from this, the robust tracking performance and accordingly the stability of the proposed control system are guaranteed by the utilization of the H∞control approach. The proposed BENN is validated and evaluated by a chaos synchronization simulation, and the overall control system by object grasping tasks through a physical robotic arm in a real-world environment. The experimental results demonstrate the superiority of the proposed control system in reference to those with single neural networks.

Image super resolution is one of the most popular topics in the field of image processing. However, most of the existing super resolution algorithms are designed for the situation where sufficient training data is available. This paper proposes a new image super resolution approach that is able to handle the situation with sparse training data, using the recently developed ANFIS (Adaptive Network based Fuzzy Inference System) interpolation technique. In particular, the training image data set is divided into different subsets. For subsets with sufficient training data, the ANFIS models are trained using standard ANFIS learning procedure, while for those with insufficient data, the models are obtained through ANFIS interpolation. In the literature, little work exists for image super resolution on sparse data. Therefore, in the experimental evaluations of this paper, the proposed approach is compared with existing super resolution methods with full data, demonstrating that this work is able to produce highly promising results.

This paper proposes a self-organizing control system for uncertain nonlinear systems. The proposed neural network is composed of a conventional brain emotional learning network (BEL) and a cerebellar model articulation controller network (CMAC). The input value of the network is feed to a BEL channel and a CMAC channel. The output of the network is generated by the comprehensive action of the two channels. The structure of the network is dynamic, using a self-organizing algorithm allows increasing or decreasing weight layers. The parameters of the proposed network are on-line tuned by the brain emotional learning rules; the updating rules of CMAC and the robust controller are derived from the Lyapunov function; in addition, stability analysis theory is used to guaranty the proposed controller's convergence. A simulated mobile robot is applied to prove the effectiveness of the proposed control system. By comparing with the performance of other neural-network-based control systems, the proposed network produces better performance.

Dynamic control, including robotic control, faces both the theoretical challenge of obtaining accurate system models and the practical difficulty of defining uncertain system bounds. To facilitate such challenges, this paper proposes a control system consisting of a novel type of fuzzy neural network and a robust compensator controller. The new fuzzy neural network is implemented by integrating a number of key components embedded in a Type-2 fuzzy cerebellar model articulation controller (CMAC) and a brain emotional learning controller (BELC) network, thereby mimicking an ideal sliding mode controller. The system inputs are fed into the neural network through a Type-2 fuzzy inference system (T2FIS), with the results subsequently piped into sensory and emotional channels which jointly produce the final outputs of the network. That is, the proposed network estimates the nonlinear equations representing the ideal sliding mode controllers using a powerful compensator controller with the support of T2FIS and BELC, guaranteeing robust tracking of the dynamics of the controlled systems. The adaptive dynamic tuning laws of the network are developed by exploiting the popular brain emotional learning rule and the Lyapunov function. The proposed system was applied to a robot manipulator and a mobile robot, demonstrating its efficacy and potential; and a comparative study with alternatives indicates a significant improvement by the proposed system in performing the intelligent dynamic control.

Fuzzy rule interpolation (FRI) facilitates approximate reasoning in fuzzy rule-based systems only with sparse knowledge available, remedying the limitation of conventional compositional rule of inference working with a dense rule base. Most of the existing FRI work assumes equal significance of the conditional attributes in the rules while performing interpolation. Recently, interesting techniques have been reported for achieving weighted interpolative reasoning. However, they are either particularly tailored to perform classification problems only or employ attribute weights that are obtained using additional information (rather than just the given rules), without integrating them with the associated FRI procedure. This paper presents a weighted rule interpolation scheme for performing prediction tasks by the use of fuzzy sparse knowledge only. The weights of rule conditional attributes are learned from a given rule base to discriminate the relative significance of each individual attribute and are integrated throughout the internal mechanism of the FRI process. This scheme is demonstrated using the popular scale and move transformation-based FRI for resolving prediction problems, systematically evaluated on 12 benchmark prediction tasks. The performance is compared with the relevant state-of-the-art FRI techniques, showing the efficacy of the proposed approach.

Fuzzy rule interpolation (FRI) is of particular significance for reasoning in the presence of insufficient knowledge or sparse rule bases. As one of the most popular FRI methods, transformation-based fuzzy rule interpolation (TFRI) works by constructing an intermediate fuzzy rule, followed by running scale and move transformations. The process of intermediate rule construction selects a user-defined number of rules closest to an observation that does not match any existing rule, using a distance metric. It relies upon heuristically computed weights to assess the contribution of individual selected rules. This process requires a move operation in an effort to force the intermediate rule to overlap with an unmatched observation, regardless of what rules are selected and how much contribution they may each make. It is, therefore, desirable to avoid this problem and also to improve the automation of rule interpolation without resorting to the user's intervention for fixing the number of closest rules. This article proposes such a novel approach to selecting a subset of rules from the sparse rule base with an embedded rule weighting scheme for the automatic assembling of the intermediate rule. Systematic comparative experimental results are provided on a range of benchmark datasets to demonstrate statistically significant improvement in the performance achieved by the proposed approach over that obtainable using conventional TFRI.

Fuzzy rule interpolation (FRI) enables sparse fuzzy rule-based systems to derive an interpolated conclusion using neighboring rules, when presented with an observation that matches none of the given rules. The efficacy of FRI has been further empowered by the recent development of weighted FRI techniques, particularly the one that introduces attribute weights of rule antecedents from the given rule base, removing the conventional assumption of antecedent attributes having equal weighting or significance. However, such work was carried out within the specific transformation-based FRI mechanism. This short paper reports the results of generalizing it through enhancing two alternative representative FRI methods. The resultant weighted FRI algorithms facilitate the individual attribute weights to be integrated throughout the corresponding procedures of the conventional unweighted methods. With systematical comparative evaluations over benchmark classification problems, it is empirically demonstrated that these algorithms work effectively and efficiently using just two nearest neighboring rules.

Sentiment analysis mines people's opinions and attitudes regarding a certain issue from source materials. Recently, it has drawn significant attention in a number of application areas. The sentiment analysis of healthcare in general and that of users' drug experience in particular could shed significant light on how to improve public health and make the right decisions. However, one of the major challenges in sentiment classification lies in the very large number of extracted features. Fuzzy-rough feature selection provides a means by which discrete or real-valued noisy data can be effectively reduced without human intervention. This paper proposes an implementation for automatic sentiment classification of drug reviews employing fuzzy rough feature selection. Experimental results demonstrate that the employment of fuzzy-rough feature selection can indeed significantly reduce the complexity of feature space and the classification run-time overheads while maintaining classification accuracy.

Dynamic fuzzy rule interpolation (D-FRI) enhances the accuracy of sparse rule-based fuzzy reasoning via efficiently exploiting fuzzy rule interpolation to produce dynamic rules. Owing to its adaptive nature in delivering a dynamic rule base, it is particularly useful for those systems which experience frequent changes. Network security is one such area where frequent changes are quite likely due to changing network conditions and traffic. Thus, D-FRI has the potential to offer an optimised and adaptive approach for improving network security. The popular Cisco Adaptive Security Appliance (ASA) Firewall is capable of monitoring and alerting a range of common threats, by baselining the traffic of a network and analysing the statistics of dropped packets. An ASA process yields a large volume of statistical information relating to certain security events. Yet, threat detection is a rudimentary function since additional intelligence is required to automate the extraction of meaningful information for alerting the users. This could be achieved using expensive automated tools offered by a third party, but doing so may unnecessarily expose an organisation to other security threats. This paper takes a different approach, presenting a DFRI-CiscoFirewall in support of automated threat detection for Cisco ASA Firewall. Through utilising threat detection statistics, the approach can customise the detection process according to organisational requirements. It performs the relative analysis of prioritised security events and is able to predict comprehensive security situations while no matching rules are available. In particular, the approach supports the creation of a dynamic rule base, derived from changing network conditions and traffic density. Its efficacy is demonstrated by experimental evaluations.

CONTEXT AND BACKGROUND: Breast cancer is one of the most common diseases threatening the human lives globally, requiring effective and early risk analysis for which learning classifiers supported with automated feature selection offer a potential robust solution. MOTIVATION: Computer aided risk analysis of breast cancer typically works with a set of extracted mammographic features which may contain significant redundancy and noise, thereby requiring technical developments to improve runtime performance in both computational efficiency and classification accuracy. HYPOTHESIS: Use of advanced feature selection methods based on multiple diagnosis criteria may lead to improved results for mammographic risk analysis. METHODS: An approach for multi-criterion based mammographic risk analysis is proposed, by adapting the recently developed multi-label fuzzy-rough feature selection mechanism. RESULTS: A system for multi-criterion mammographic risk analysis is implemented with the aid of multi-label fuzzy-rough feature selection and its performance is positively verified experimentally, in comparison with representative popular mechanisms. CONCLUSIONS: The novel approach for mammographic risk analysis based on multiple criteria helps improve classification accuracy using selected informative features, without suffering from the redundancy caused by such complex criteria, with the implemented system demonstrating practical efficacy.

Network intrusion detection systems (IDSs) dynamically monitor communication events on a network, and decide whether any event is symptomatic of an attack or constitutes a legitimate use of the system. They have become an indispensable component of security infrastructure, e.g., to detect threats before widespread damage takes place. A variety of approaches have been proposed to design IDSs, including fuzzy rule-based techniques that offer advantages such as tolerance of noisy and imprecise data. In particular, fuzzy rules can be highly interpretable and trackable if the underlying fuzzy sets are predefined, directly reflecting domain expertise. This paper proposes such an approach to generate a set of weighted fuzzy rules for building effective IDSs, where the rule weights are optimised by Particle Swarm Optimisation without affecting the underlying predefined fuzzy sets. Experiments are performed on benchmark IDS datasets with comparison to alternative systems built with popular machine learning methods.

This paper presents a novel fuzzy rule-based interpolative reasoning system for mammographic mass shape classification that is interpretable to medical professionals. In particular, a feature ranking-guided fuzzy rule interpolation (FRI) method is embedded in the proposed system to make inference possible given a sparse rule base, which may occur in dealing with insufficient mammographic image data (and indeed in coping with many other computer-aided medical diagnostic problems). The rule base for inference is learned from a set of labelled morphological features which are extracted from mass shapes. A classical FRI mechanism is integrated with a procedure for feature selection to score the individual rule antecedents in the inducted sparse rule base for more accurate interpolative reasoning. The work is evaluated on a real-world mammographic image data base with promising results, demonstrating the efficacy of the proposed fuzzy rule-based interpolative classification system.

Conventional approaches of robotic writing of Chinese character strokes often suffer from limited font generation methods, and thus the writing results often lack of diversity. This has seriously restricted the high quality writing ability of robots. This paper proposes a generative adversarial nets-based calligraphic robotic framework, which enables a robot to learn writing fundamental Chinese strokes with rich diversity and good originality. In particular, the framework considers the learning process of robotic writing as an adversarial procedure which is implemented by three interactive modules including a stroke generation module, a stroke discriminative module and a training module. Noting that the stroke generative module included in the conventional generative adversarial nets cannot solve the non-differentiable problem, the policy gradient commonly used in reinforcement learning is thus adapted in this work to train the generative module by regarding the outputs from the discriminative module as rewards. Experimental results demonstrate that the proposed framework allows a calligraphic robot to successfully write fundamental Chinese strokes with good quality in various styles. The experiment also suggests the proposed approach can achieve human-level stroke writing quality without the requirement of a performance evaluation system. This approach therefore significantly boosts the robotic autonomous creation ability.

This paper proposes a novel approach for building transparent knowledge-based systems by generating accurate and interpretable fuzzy rules. The learning mechanism reported here induces fuzzy rules via making use of only predefined fuzzy labels that reflect prescribed notations and domain expertise, thereby ensuring transparency in the knowledge model adopted for problem solving. It works by mapping every coarsely learned crisp production rule in the knowledge base onto a set of potentially useful fuzzy rules, which serves as an initial step towards an intuitive technique for similarity-based rule generalisation. This is followed by a procedure that locally selects a compact subset of the emerging fuzzy rules, so that the resulting subset collectively generalises the underlying original crisp rule. The outcome of this local procedure forms the input to a global genetic search process, which seeks for a trade-off between accuracy and complexity of the eventually induced fuzzy rule base while maintaining transparency. Systematic experimental results are provided to demonstrate that the induced fuzzy knowledge base is of high performance and interpretability.

Fuzzy rule interpolation (FRI) makes inference possible when dealing with a sparse and imprecise rule base. However, the rule antecedents are commonly assumed to be of equal significance in most FRI approaches in the implementation of interpolation. This may lead to a poor performance of interpolative reasoning due to inaccurate or incorrect interpolated results. In order to improve the accuracy by minimising the disadvantage of the equal significance assumption, this paper presents a novel inference system where an information gain (IG)-guided fuzzy rule interpolation method is embedded. In particular, the rule antecedents in FRI are weighted using IG to evaluate the relative importance given the consequent for decision making. The computation of antecedent weights is enabled by introducing an innovative reverse engineering process that artificially converts fuzzy rules into training samples. The antecedent weighting scheme is integrated with scale and move transformation-based interpolation (though other FRI techniques may be improved in the same manner). An illustrative example is used to demonstrate the execution of the proposed approach, while systematic comparative experimental studies are reported to demonstrate the potential of the proposed work.

Identification of hidden relationships between domain attributes from different data sources is of great practical significance and forms an emerging field in data mining. However, currently there seldom exist any systematic methods that can effectively handle this problem, especially when dealing with imprecisely described associations. In this paper, a novel data-driven approach for inter-variable correlation prediction is proposed by exploiting the concept of connected-triples. The work is implemented with the use of fuzzy logic. Through the exploitation of link strength measurements and fuzzy inference, the job of detecting similar or related variables can be accomplished via examining link relation patterns within and across different data sources. Empirical evaluation results are discussed, revealing the potential of the proposed work in predicting interesting attribute relations, while involving simple computation mechanisms.

The k nearest-neighbour ( k NN) algorithm has enjoyed much attention since its inception as an intuitive and effective classification method. Many further developments of k NN have been reported such as those integrated with fuzzy sets, rough sets, and evolutionary computation. In particular, the fuzzy and rough modifications of k NN have shown significant enhancement in performance. This paper presents another significant improvement, leading to a multi-functional nearest-neighbour (MFNN) approach which is conceptually simple to understand. It employs an aggregation of fuzzy similarity relations and class memberships in playing the critical role of decision qualifier to perform the task of classification. The new method offers important adaptivity in dealing with different classification problems by nearest-neighbour classifiers, due to the large and variable choice of available aggregation methods and similarity metrics. This flexibility allows the proposed approach to be implemented in a variety of forms. Both theoretical analysis and empirical evaluation demonstrate that conventional k NN and fuzzy nearest neighbour, as well as two recently developed fuzzy-rough nearest-neighbour algorithms can be considered as special cases of MFNN. Experimental results also confirm that the proposed approach works effectively and generally outperforms many state-of-the-art techniques.

Using fuzzy rule interpolation (FRI) interpolative reasoning can be effectively performed with a sparse rule base where a given system observation does not match any fuzzy rules. While offering a potentially powerful inference mechanism, in the current literature, typical representation of fuzzy rules in FRI assumes that all attributes in the rules are of equal significance in deriving the consequents. This is a strong assumption in practical applications, thereby, often leading to less accurate interpolated results. To address this challenging problem, this paper employs feature selection (FS) techniques to adjudge the relative significance of individual attributes and therefore, to differentiate the contributions of the rule antecedents and their impact upon FRI. This is feasible because FS provides a readily adaptable mechanism for evaluating and ranking attributes, being capable of selecting more informative features. Without requiring any acquisition of real observations, based on the originally given sparse rule base, the individual scores are computed using a set of training samples that are artificially created from the rule base through an innovative reverse engineering procedure. The attribute scores are integrated within the popular scale and move transformation-based FRI algorithm (while other FRI approaches may be similarly extended following the same idea), forming a novel method for attribute ranking-supported fuzzy interpolative reasoning. The efficacy and robustness of the proposed approach is verified through systematic experimental examinations in comparison with the original FRI technique over a range of benchmark classification problems while utilizing different FS methods. A specific and important outcome that is supported by attribute ranking, only two (i.e., the least number of) nearest adjacent rules are required to perform accurate interpolative reasoning, avoiding the need of searching for and computing with multiple rules beyond the immediate neighborhood of a given observation.

Classifier ensembles form an important approach to improving classification performance. As such, there have been different proposals made in the literature that provide a range of means to construct and aggregate classifier ensembles. However, the resulting systems may contain unreliable members with false or biased judgements in the ensemble. The removal of unreliable members is necessary to optimise the overall performance of such systems. Smaller ensembles involving reduced ensemble members also helps relax the requirement of computational memory, thereby strengthening the system's run-time efficiency. To differentiate the potential contributions of different ensemble members while reducing the adverse impact of any unreliable judgement upon the system, a nearest neighbour-based reliability measure is incorporated into the process of classifier ensemble selection. In particular, reliabilities of selected ensemble members are perceived as a stress function, from which argument-dependent weights are heuristically generated for final aggregated decision. Experimental investigations are carried out, demonstrating the efficacy of the proposed approach, where fuzzy classifiers are utilised as base members of the emerging ensemble.

Fuzzy rule interpolation (FRI) provides an alternative means to make inference with a sparse rule base, rather than directly resulting in failed reasoning when no rules can be fired for an input observation. However, existing approaches to FRI typically assume that rule antecedents are of equal significance in the implementation of interpolation, thereby often leading to less accurate interpolated results. Having taken notice of feature selection (FS) techniques being capable of selecting (subsets of) informative features, providing a mechanism of evaluating and ranking features, this work employs FS to score the individual rule antecedents in a given rule base. In particular, the computation of individual scores is enabled by the introduction of an innovative reverse engineering technique that artificially creates a set of training samples from a given sparse rule base. The antecedent scores are integrated within the scale and move transformation-based FRI algorithm (though other FRI approaches may employ the same idea), forming a novel feature ranking-guided FRI method. The work is systematically examined, by utilising six different FS techniques and comparing over eight benchmark classification problems, demonstrating improved classification performance.

Dynamic fuzzy rule interpolation (D-FRI) consists of functionalities of fuzzy rule interpolation and dynamically refinement of the fuzzy rule base. It can be integrated with any fuzzy intelligent system to extend the system's capabilities in addition to its normal fuzzy reasoning. Systems security is one of the areas that require dynamic monitoring due to the nature of possible threats; static rule-based systems cannot cover all reoriented security threats accurately in the long run. D-FRI provides a possible solution to such problems, potentially making various security tools (e.g., those for firewall, intrusion detection and traffic analysis) more effective. As a particular application, this paper exploits D-FRI to dynamically support Microsoft Windows Firewall, resulting in a robust system named D-FRI-WinFirewall. Given the general utility of Windows Firewall, the impact of this work is far-reaching. The work reported here focusses on the monitoring and prevention of denial of service (DoS) attacks, which is not possible by utilising the standard Windows Firewall alone. In particular, two sub-systems are designed, implemented and tested within D-FRI-WinFirewall, with an effort to detect and prevent two serious types of DoS attack: ICMP DoS attack and UDP DoS attack, leading the Windows Firewall to outperform popular and expensive firewalls, which are yet unable to handle DoS attacks.

The interaction between features, or attributes, of a dataset forms a major topic in machine learning and data mining. In particular, a wide range of methods have been established for feature selection, ranking, and grouping. Amongst these, fuzzy rough set based feature selection (FRFS) has been shown to be highly effective at reducing dimensionality for real-valued datasets while retaining attribute semantics. In fuzzy rough sets, the concept of crisp equivalence classes is extended by fuzzy similarity relations, and real-valued similarity measures can be captured between data instances in terms of their attribute values. Therefore, it is desirable to study the aggregation of fuzzy similarity relations to reflect the interactions between attributes. This paper presents an approach that employs OWA aggregation of fuzzy similarity relations to better perform FRFS. A high degree of modelling flexibility is provided by choosing the stress function in OWA. Experimental studies demonstrate that through using different stress functions, different features may be selected; and that given an appropriate stress function, the quality of selected features can improve over that achievable by the state-of-the-art FRFS, in performing classification tasks.

This paper proposes a new approach for single frame image super resolution using multiple ANFIS (Adaptive Network-based Fuzzy Inference System) mappings. It presents an implemented learning system that captures the relationship between a low resolution (LR) image patch space and a high resolution (HR) one given an external image database. In particular, a collected large number of LR and HR image patch pairs are divided into different groups with a clustering method. For each clustered group of the training samples, an ANFIS mapping is learned for super resolution (SR). The non-local means filter is subsequently employed to suppress the displeasing artefacts of the resulting reconstructed HR image. The proposed approach is evaluated on a range of natural images and compared with a number of existing state-of-the-art SR algorithms, demonstrating its effectiveness.

Journal impact indicators are widely accepted as possible measurements of academic journal quality. However, much debate has recently surrounded their use, and alternative journal impact evaluation techniques are desirable. Aggregation of multiple indicators offers a promising method to produce a more robust ranking result, avoiding the possible bias caused by the use of a single impact indicator. In this paper, fuzzy aggregation and fuzzy clustering, especially the ordered weighted averaging (OWA) operators are exploited to aggregate the quality scores of academic journals that are obtained from different impact indicators. Also, a novel method for linguistic term-based fuzzy cluster grouping is proposed to rank academic journals. The paper allows for the construction of distinctive fuzzy clusters of academic journals on the basis of their performance with respect to different journal impact indicators, which may be subsequently combined via the use of the OWA operators. Journals are ranked in relation to their memberships in the resulting combined fuzzy clusters. In particular, the nearest-neighbor guided aggregation operators are adopted to characterize the reliability of the indicators, and the fuzzy clustering mechanism is utilized to enhance the interpretability of the underlying ranking procedure. The ranking results of academic journals from six subjects are systematically compared with the outlet ranking used by the Excellence in Research for Australia, demonstrating the significant potential of the proposed approach.

The prediction of student academic performance is important to both educational institutions and students themselves for a variety of reasons. However, previous techniques often consider only past numeric data for prediction, whereas others overuse different types of indicative attribute, leading to the creation of complicated predicting methods whose results are difficult to interpret. This paper proposes a novel approach to predicting student final period grade, using attributes related to student past academic records and attributes of normal study behaviour, which are readily obtainable and easily interpretable. The proposed approach works by employing fuzzy clustering and multi-variable regression within an integrated framework, which also includes an offset value mechanism to support the use of attributes that are related to normal student study behaviour. Comparative experimental investigations are carried out, demonstrating the potential of the proposed work in producing more accurate results.

The use of fuzzy quantifiers to modify the fuzzy linguistic terms in fuzzy models helps build fuzzy systems in a more natural way, by capturing finer pieces of information embedded in the training data. This paper presents a practical approach for the acquisition of fuzzy production rules with quantifiers, based on a class-dependent simultaneous rule learning strategy where each class is associated with a subset of descriptive rules. It is implemented by particle swam optimisation. The performance of the learned fuzzy rules with and without fuzzy quantifiers is evaluated on various UCI benchmark data sets, in comparison to popular alternative rule based learning classifiers. Experimental results demonstrate that rule bases generated by the proposed approach indeed boost classification performance as compared to those involving no fuzzy quantifiers, with at least competitive performance to the alternative learning classifiers.

This paper presents a novel application of interpolation in supporting fuzzy image classification. The recently introduced Deep Spatio-Temporal Inference Network (DeSTIN) is employed to carry out limited original feature extraction. A simple but effective linear interpolation is then used to artificially increase the dimensionality of the extracted feature sets for accurate classification, without incurring heavy computational cost. In particular, Fuzzy-Rough Nearest Neighbour (FRNN) and Fuzzy Ownership Nearest Neighbour (FRNN-O) are each utilised for image classification. The work is tested against the popular MNIST dataset of handwritten digits [1]. Experimental results indicate that the proposed approach is highly promising.

Aggregation operators are useful tools which summarise multiple inputs to a single output. In practice, inputs to such operators are variables which represent different criteria, measurements, or opinions from experts. In this paper, a nearest neighbour-guided induced OWA operator, abbreviated as kNN-IOWA, is proposed as a special case of the generic induced OWA where the input arguments are ordered by the average distances to their k nearest neighbours. The weighting vectors in kNN-IOWA are defined, which are used to interpret the overall behaviour of the operator's reliability. kNN-IOWA is applied for building aggregated fuzzy relations between academic journals, based on their indicator scores. It combines the similarities between academic journals to assess their performance with respect to different journal impact indicators. The work is compared against different types of aggregation operator and tested on six bibliometric datasets. The results of experimental evaluation demonstrate that kNN-IOWA outperforms other aggregation operators in terms of standard accuracy and within-1 accuracy. The proposed method also exhibits the advantages of being more intuitive and interpretable.

Cluster ensemble offers an effective approach for aggregating multiple clustering results in order to improve the overall clustering robustness and stability. It also helps improve accuracy by combing clustering results from component methods that utilise different parameters (e.g., number of clusters), avoiding the need for carefully pre-setting the values of such parameters in a single clustering process. Since founded, many topics regarding cluster ensemble have been proposed and promising results gained. These include the generation of ensemble members and consensus of ensemble members. In this paper, link-based consensus methods for the ensemble of fuzzy c-means are proposed. Different from traditional clustering techniques, the clusters which are generated by fuzzy c-means are fuzzy sets. The proposed methods therefore employ a fuzzy graph to represent the relationships between component clusters upon which to derive the final ensemble clustering results. Using various benchmark datasets, the proposed methods are tested against typical traditional methods. The experimental results demonstrate that the proposed fuzzy-link-based clustering ensemble approach generally outperforms the others in terms of accuracy.

This paper presents a novel approach for image classification, by integrating the concepts of deep machine learning and feature interpolation. In particular, a recently introduced learning architecture, the Deep Spatio-Temporal Inference Network (DeSTIN) [1] is employed to perform feature extraction for support vector machine (SVM) based image classification. Linear interpolation and Newton polynomial interpolation are each applied to support the classification. This approach converts feature sets of an originally low-dimensionality into those of a significantly higher dimensionality while gaining overall computational simplification. The work is tested against the popular MNIST dataset of handwritten digits [2]. Experimental results indicate that the proposed approach is highly promising.

The ability of pointing is recognised as an essential skill of a robot in its communication and social interaction. This paper introduces a developmental learning approach to robotic pointing, by exploiting the interactions between a human and a robot. The approach is inspired through observing the process of human infant development. It works by first applying a reinforcement learning algorithm to guide the robot to create attempt movements towards a salient object that is out of the robot’s initial reachable space. Through such movements, a human demonstrator is able to understand the robot desires to touch the target and consequently, to assist the robot to eventually reach the object successfully. The human–robot interaction helps establish the understanding of pointing gestures in the perception of both the human and the robot. From this, the robot can collect the successful pointing gestures in an effort to learn how to interact with humans. Developmental constraints are utilised to drive the entire learning procedure. The work is supported by experimental evaluation, demonstrating that the proposed approach can lead the robot to gradually gain the desirable pointing ability. It also allows that the resulting robot system exhibits similar developmental progress and features as with human infants.

Fuzzy similarity relations form the basis for many developments and applications of fuzzy systems. Measures of fuzzy similarity have been proposed in the literature for comparing objects. In this paper, aggregated fuzzy relations are generated between academic journals to compare their performance with respect to different journal impact indicators. In particular, various indicators may be employed to construct several distinctive fuzzy similarity relations, which may be subsequently combined via the use of the Ordered Weighted Average (OWA) operator. This proposed aggregated measure preserves reflexivity and symmetry, with T-transitivity conditionally preserved if appropriate weighting vectors are selected. Different similarity measures and weighting vectors are compared for the task of journal clustering, in an effort to estimate the ranking of academic journals. The results of experimental evaluation demonstrate that by using OWA-aggregated relations, simple techniques such as C-means can perform well in terms of standard accuracy and within-1 accuracy. The proposed method also exhibits the advantages of being more intuitive and interpretable.

This paper presents a study on rock texture image classification using support vector machines (and also K-nearest neighbours and decision trees) with the aid of feature selection techniques. It offers both unsupervised and supervised methods for feature selection, based on data reliability and information gain ranking respectively. Following this approach, the conventional classifiers which are sensitive to the dimensionality of feature patterns, become effective on classification of images whose pattern representation may otherwise involve a large number of features. The work is successfully applied to complex images. Classifiers built using features selected by either of these methods generally outperform their counterparts that employ the full set of original features which has a dimensionality several folds higher than that of the selected feature subset. This is confirmed by systematic experimental investigations. This study therefore, helps to accomplish challenging image classification tasks effectively and efficiently. In particular, the approach retains the underlying semantics of a selected feature subset. This is very important to ensure that the classification results are understandable by the user.

Fuzzy-rough sets play an important role in dealing with imprecision and uncertainty for discrete and real-valued or noisy data. However, there are some problems associated with the approach from both theoretical and practical viewpoints. These problems have motivated the hybridisation of fuzzy-rough sets with kernel methods. Existing work which hybridises fuzzy-rough sets and kernel methods employs a constraint that enforces the transitivity of the fuzzy T-norm operation. In this paper, such a constraint is relaxed and a new kernel-based fuzzy-rough set approach is introduced. Based on this, novel kernel-based fuzzy-rough nearest-neighbour algorithms are proposed. The work is supported by experimental evaluation, which shows that the new kernel-based methods offer improvements over the existing fuzzy-rough nearest neighbour classifiers. The abstract goes here.

The measure of data reliability has recently proven useful for a number of data analysis tasks. This paper extends the underlying metric to a new problem of soft subspace clustering. The concept of subspace clustering has been increasingly recognized as an effective alternative to conventional algorithms (which search for clusters without differentiating the significance of different data attributes). While a large number of crisp subspace approaches have been proposed, only a handful of soft counterparts are developed with the common goal of acquiring the optimal cluster-specific dimension weights. Most soft subspace clustering methods work based on the exploitation of k-means and greatly rely on the iteratively disclosed cluster centers for the determination of local weights. Unlike such wrapper techniques, this paper presents a filter approach which is efficient and generally applicable to different types of clustering. Systematical experimental evaluations have been carried out over a collection of published gene expression data sets. The results demonstrate that the reliability-based methods generally enhance their corresponding baseline models and outperform several well-known subspace clustering algorithms.

This paper presents a novel application of support vector machine (SVM) based classifiers for Mars terrain image classification. SVMs are applied in conjunction with information gain ranking (IGR) that allows the induction of informative feature subsets from sample descriptions of feature vectors of a higher dimensionality. Such an integrated use of IGR and SVMs helps to enhance the effectiveness and efficiency of the classifiers, minimizing redundant and noisy features. This work is supported with comparative studies - the resultant SVM-based classifiers generally outperform MLP and KNN-based classifiers and those which use PCA-returned features.

This paper presents a novel study of the classification of large-scale Mars McMurdo panorama image. Three dimensionality reduction techniques, based on fuzzy-rough sets, information gain ranking, and principal component analysis respectively, are each applied to this complicated image data set to support learning effective classifiers. The work allows the induction of low-dimensional feature subsets from feature patterns of a much higher dimensionality. To facilitate comparative investigations, two types of image classifier are employed here, namely multi-layer perceptrons and K-nearest neighbors. Experimental results demonstrate that feature selection helps to increase the classification efficiency by requiring considerably less features, while improving the classification accuracy by minimizing redundant and noisy features. This is of particular significance for on-board image classification in future Mars rover missions.

This paper presents a methodological approach for developing image classifiers that work by exploiting the technical potential of both fuzzy-rough feature selection and neural network-based classification. The use of fuzzy-rough feature selection allows the induction of low-dimensionality feature sets from sample descriptions of real-valued feature patterns of a (typically much) higher dimensionality. The employment of a neural network trained using the induced subset of features ensures the runtime classification performance. The reduction of feature sets reduces the sensitivity of such a neural network-based classifier to its structural complexity. It also minimises the impact of feature measurement noise to the classification accuracy. This work is evaluated by applying the approach to classifying real medical cell images, supported with comparative studies.

This paper presents a novel approach to the analysis and classification of tissue section images of human resistance arteries. Real tissue images are modelled using directional fractal dimensions and a multi-layer feedforward neural network is adopted to perform the classification task. This approach has been applied to a large database of images. Simulation results show that modelling cell images with directional fractal dimensions allows the capture of differentiating features not only between normal and abnormal cells but also between the categories within such cells. Directional fractal features entail better discrimination than multi-resolution ones.

This paper presents a novel application of simultaneous autoregressive models to the synthesis of magnetic material images. The effects of using either symmetric or non-symmetric neighbour sets upon the visual and statistical properties of the resulting synthesised images are investigated. The use of a neighbour set whose shape corresponds to the orientations and coarseness of the texture allows the generation of synthetic images of good quality. Also, the size of such a neighbour set is usually smaller than that of the symmetric set required to reach similar modelling accuracy, thereby minimising the computational effort.

This paper presents a novel approach to adapting the weights of single-layer perceptron based communication channel equalisers, using the log-likelihood cost function and the resulting least-mean-log-likelihood adaptive algorithm. It provides promising theoretical and experimental results, illustrating that developing equalisers following this approach allows faster convergence, lower estimation error and better bit error rate performance in comparison with the traditional Rosenblatt (1962) and back-propagation algorithms.