According to the traditional Chinese medicine composite constitution theory, body constitution recognition is modeled as a unique task of multi-label problems using tongue images. Although MLP-like architecture is one of the base models except for CNN-based, Transformer-based, and MLP-like-based for computer vision tasks, the performance of the MLP-like-based architecture for the multi-label BCR task needs to expand for multi-label BCR task with a specific module using tongue images. Thus, a novel dual transform MLP-like (DT-MLP) model is designed, in which discrete Fourier transform and chaos transform are used for feature extraction of their branches. Besides, a new multi-label tongue body constitution dataset is constructed. The results demonstrate that the proposed DT-MLP outperforms other state-of-the-art MLP-like models on mAP, AUC, and Acc, respectively.

Facial Expression Recognition (FER) has achieved significant success in recent years due to the rise of deep learning. Meanwhile, latent semantic information is crucial for recognizing facial expressions with subtle differences. Inspired by inconsistencies in learning intensity across different layers of deep learning networks — where shallow-layer features lack generalization and task relevance compared to deep-layer features — we propose a novel Hierarchical Semantic Transfer (HST) method. This method uses attention maps from deep-layer features to regularize the model, enabling it to effectively capture the necessary semantic information and align the latent semantics between shallow and deep features, while ignoring irrelevant noise. Furthermore, to address the issue of class imbalance present in FER datasets, we introduce a Batch-wise Similarity Attention (BSA) mechanism to learn the similarity relationships between features of different samples. Extensive experiments on different datasets and backbones demonstrate that HSBS can work together to significantly improve model performance and achieve state-of-the- art results on those FER datasets.

Online social networks serve as an essential platform for human behavior, making the exploration of their impact on epidemic spread a significant issue. This study utilizes the research methodology of propagation dynamics to quantify the interaction between information dissemination in social networks and virus diffusion in physical networks. The double-layer dynamic model was constructed, introducing the concept of dynamic transmission probabilities. The dynamic infection probability of disease transmission incorporates the information immunization effect, where wide-spread information dissemination leads to spontaneous immune behavior among the population. The dynamic transmission probability of information spread incorporates the fear effect, where a surge in mortality rates triggers exponential information dissemination. Mathematical derivations, parameter analyses, and simulation experiments were initially conducted. Subsequently, the differential equations of the double-layer network dynamic model were used to propose the new double-layer dynamic module. Thus, this article proposes a novel epidemic forecasting framework called physics-informed graph neural network (PIGNN). Extensive experiments show that our proposed PIGNN model outperforms the state-of-the-art method. Meanwhile, the double-layer network dynamic model was demonstrated to capture real-world phenomena vividly.

The theory of "three-stage prevention" in view of the body constitution is the key technology of modern Chinese medicine for "Preventive Treatment of Diseases". In particular, automated body constitution recognition (BCR) is an integral part of intelligent Traditional Chinese Medicine (TCM), which is extremely valuable for disease prevention and diagnosis. Actually, BCR is a challenging multi-label recognition task by the TCM composite constitution theory. First, two new databases are constructed, one is a multi-label facial body constitution (MFBC), and another is a multi-label tongue body constitution (MTBC). Second, a novel MLP-like architecture, named Chaos-MLP, is designed for the BCR task, which interacts with the channel chaotic features of extracted medical images and fuses them with the width and height channel direction features, respectively. Notably, the chaotic transform can enhance the distinguishability of extracted features from the medical images. Moreover, we propose a binary center cognitive gravity loss (BCCGL) to enhance the learning ability of the Chaos-MLP for unbalanced body constitution labels. Our proposed method shows superior performance on both MFBC and MTBC datasets than other state-of-the-art (SOTA) MLP-like networks and a vision graph-based neural network (VGNN), which include Wave-MLP, Cycle-MLP, Vip, and Active-MLP.

Cross-Corpus Speech Emotion Recognition (SER) aims to identify human emotions from speech across different speakers and languages. Previous work engaged in extracting the domain-invariant features among individual samples that are most relevant to emotions, ignoring rich relationships between speech instances, which are also significant factors that strongly influence the sentiments. To explore those potential relationships across multiple corpora, we introduce a novel cross-corpus SER architecture with speech relationship learning. Specifically, during training, we employ the attention mechanism on the entire input batch, embedding the sample-level similar features in emotion space into new representations. Furthermore, a dual discriminator structure is proposed for improving the similarity calculation performance through adversarial training, and a domain-wise shared classifier with batch label smoothing strategy is proposed to enhance the network generalization ability. Experiments on the CASIA, EMODB and SAVEE datasets have demonstrated that the proposed method outperforms the state-of-the-art cross-corpus SER methods.

Facial expression recognition (FER) has a wide range of applications, including interactive gaming, healthcare, security, and human‐computer interaction systems. Despite the impressive performance of FER based on deep learning methods, it remains challenging in real‐world scenarios due to uncontrolled factors such as varying lighting conditions, face occlusion, and pose variations. In contrast, humans are able to categorize objects based on both their inherent characteristics and the surrounding environment from a cognitive standpoint, utilizing concepts such as cognitive relativity. Modeling the cognitive relativity laws to learn cognitive features as feature augmentation may improve the performance of deep learning models for FER. Therefore, we propose a cognitive feature learning framework to learn cognitive features as complementary for FER, which consists of Relative Transformation module (AFRT) and Graph Convolutional Network module (AFGCN). AFRT explicitly creates cognitive relative features that reflect the position relationships between the samples based on human cognitive relativity, and AFGCN implicitly learns the interaction features between expressions as feature augmentation to improve the classification performance of FER. Extensive experimental results on three public datasets show the universality and effectiveness of the proposed method.

Video-based facial expression recognition (FER) is a challenging task due to the dynamic emotional changes with variant frames in video sequences. This paper proposes a novel coarse-fine aware network with static-dynamic adaptation (CFAN-SDA) for in-the wild video-based FER. From coarse to fine, our method leverages cross-domain static FER database to boost video-based FER performance, and then explore hierarchical spatial-temporal feature learning. Specifically, different from existing methods, we design a static-dynamic adaptation learning to explore the knowledge transfer from labeled static images to unlabeled frames of video, which captures the features of coarse-grained emotion to find those important expression-related frames. Furthermore, we present hierarchical spatial-temporal transformers to better learn features of fine-grained expression, which consist of multi-view spatial transformer and frame-clip temporal transformer. The former captures multi-view spatial regions information from global to local, and the latter achieves cross-frame and cross-clip temporal interaction to select the key frame-level and clip-level multi-scale temporal information for fusing. Extensive experimental results on dynamic FER databases indicate that CFAN-SDA achieves superior performance compared to the state-of-the-art models.

Unsupervised domain adaptation (UDA) aims to estimate a transferable model for unlabeled target domains by exploiting labeled source data. Optimal Transport (OT) based methods have recently been proven to be a promising solution for UDA with a solid theoretical foundation and competitive performance. However, most of these methods solely focus on domain-level OT alignment by leveraging the geometry of domains for domain-invariant features based on the global embeddings of images. However, global representations of images may destroy image structure, leading to the loss of local details that offer category-discriminative information. This study proposes an end-to-end Deep Hierarchical Optimal Transport method (DeepHOT), which aims to learn both domain-invariant and category-discriminative representations by mining hierarchical structural relations among domains. The main idea is to incorporate a domain-level OT and image-level OT into a unified OT framework, hierarchical optimal transport, to model the underlying geometry in both domain space and image space. In DeepHOT framework, an image-level OT serves as the ground distance metric for the domain-level OT, leading to the hierarchical structural distance. Compared with the ground distance of the conventional domain-level OT, the image-level OT captures structural associations among local regions of images that are beneficial to classification. In this way, DeepHOT, a unified OT framework, not only aligns domains by domain-level OT, but also enhances the discriminative power through image-level OT. Moreover, to overcome the limitation of high computational complexity, we propose a robust and efficient implementation of DeepHOT by approximating origin OT with sliced Wasserstein distance in image-level OT and accomplishing the mini-batch unbalanced domain-level OT. Extensive experiments show the superiority of DeepHOT in several benchmark datasets. The code will be released on GitHub.

Cross-domain facial expression recognition is confronted by the problem of the large distribution discrepancy and samples inconsistencies between the source domain and target domain. To solve this problem, we propose a cross-domain sample relationship learning (CSRL) method that explores useful intrinsic sample relationships of two domains to narrow the domain discrepancy. Specifically, during the training stage, we first design inter-domain sample transformers to explore the sample similarity relationships between the source and target domains, and then deploy intra-domain sample transformers to capture the internal similar structure of the samples in each domain. Thus dual sample relationships can be learned to align the cross-domain similar samples and preserve the domain-specific information, which can facilitate both the inter-domain invariant features and intra-domain invariant features learning. Subsequently, we design a joint alignment strategy by simultaneously deploying the feature distribution alignment and cross-domain sample relationship learning. Thus, both local similar samples and global domain distribution of two domains can be well aligned to enhance the generalization ability of the model. Experimental results on several benchmark databases show the superiority of CSRL over some state-of-the-art methods.

Speech emotion recognition has lots of applications such as human‐computer interaction and health management. The current methods are challenged with the problems of fuzzy decision boundary and imbalance between difficult and easy samples in the training data. This paper first proposes an additive angle penalty focus loss function (APFL), which strictly refines the fuzzy decision boundary by introducing angle penalty factors to improve the compactness within the class and enlarge the distance between classes. It also assigns the larger loss to difficult samples to make the model pay more attention to them, as they are easily misclassified. Simultaneously, due to the lack of training samples, the framework of multimodal and multitask learning with APFL is further proposed, which extracts spectrogram features by deep neural network, text features by the pretrained language model, and audio features by the pretrained sound model. It uses the gender recognition as an auxiliary task. The experimental results verify the effectiveness of the proposed loss function and framework.

Facial expression recognition (FER) becomes more challenging in the wild due to unconstrained conditions, such as the different illumination, pose changes, and occlusion of the face. Current FER methods deploy the attention mechanism in deep neural networks to improve the performance. However, these models only capture the limited attention features and relationships. Thus this paper proposes a novel FER framework called multi-relations aware network (MRAN), which can focus on global and local attention features and learn the multi-level relationships among local regions, between global-local features and among different samples, to obtain efficient emotional features. Specifically, our method first imposes the spatial attention on both the whole face and local regions to simultaneously learn the global and local salient features. After that, a region relation transformer is deployed to capture the internal structure among local facial regions, and a global-local relation transformer is designed to learn the fusion relations between global features and local features for different facial expressions. Subsequently, a sample relation transformer is deployed to focus on intrinsic similarity relationship among training samples, which promotes invariant feature learning for each expression. Finally, a joint optimization strategy is designed to efficiently optimize the model. The conducted experimental results on in-the-wild databases show that our method obtains the superior performance compared to some state-of-the-art models.

Medical images such as facial and tongue images have been widely used for intelligence-assisted diagnosis, which can be regarded as the multi-label classification task for disease location (DL) and disease nature (DN) of biomedical images. Compared with complicated convolutional neural networks and Transformers for this task, recent MLP-like architectures are not only simple and less computationally expensive, but also have stronger generalization capabilities. However, MLP-like models require better input features from the image. Thus, this study proposes a novel convolution complex transformation MLP-like (CCT-MLP) model for the multi-label DL and DN recognition task for facial and tongue images. Notably, the convolutional Tokenizer and multiple convolutional layers are first used to extract the better shallow features from input biomedical images to make up for the loss of spatial information obtained by the simple MLP structure. Subsequently, the Channel-MLP architecture with complex transformations is used to extract deep-level contextual features. In this way, multi-channel features are extracted and mixed to perform the multi-label classification of the input biomedical images. Experimental results on our constructed multi-label facial and tongue image datasets demonstrate that our method outperforms existing methods in terms of both accuracy (Acc) and mean average precision (mAP).

Despite the tremendous success in computer vision, deep convolutional networks suffer from serious computation costs and redundancies. Although previous works address that by enhancing the diversities of filters, they have not considered the complementarity and the completeness of the internal convolutional structure. To respond to this problem, we propose a novel inner-imaging (InI) architecture, which allows relationships between channels to meet the above requirement. Specifically, we organize the channel signal points in groups using convolutional kernels to model both the intragroup and intergroup relationships simultaneously. A convolutional filter is a powerful tool for modeling spatial relations and organizing grouped signals, so the proposed methods map the channel signals onto a pseudoimage, like putting a lens into the internal convolution structure. Consequently, not only is the diversity of channels increased but also the complementarity and completeness can be explicitly enhanced. The proposed architecture is lightweight and easy to be implement. It provides an efficient self-organization strategy for convolutional networks to improve their performance. Extensive experiments are conducted on multiple benchmark datasets, including CIFAR, SVHN, and ImageNet. Experimental results verify the effectiveness of the InI mechanism with the most popular convolutional networks as the backbones.

Supervised machine learning has several drawbacks that make it difficult to use in many situations. Drawbacks include heavy reliance on massive training data, limited generalizability, and poor expressiveness of high-level semantics. Low-shot Learning attempts to address these drawbacks. Low-shot learning allows the model to obtain good predictive power with very little or no training data, where structured knowledge plays a key role as a high-level semantic representation of human. This article will review the fundamental factors of low-shot learning technologies, with a focus on the operation of structured knowledge under different low-shot conditions. We also introduce other techniques relevant to low-shot learning. Finally, we point out the limitations of low-shot learning, the prospects and gaps of industrial applications, and future research directions.

Physical signs of patients indicate crucial evidence for diagnosing both location and nature of the disease, where there is a sequential relationship between the two tasks. Thus their joint learning can utilize intrinsic association by transferring related knowledge across relevant tasks. Choosing the right time to transfer is a critical problem for joint learning. However, how to dynamically adjust when tasks interact to capture the right time for transferring related knowledge is still an open issue. To this end, we propose a Task-Coupling Elastic Learning (TCEL) framework to model the task relatedness for classifying disease-location and disease-nature based on physical sign images. The main idea is to dynamically transfer relevant knowledge by progressively shifting task-coupling from loose to tight during the multi-stage training. In the early stage of training, we relax the constraints of modeling relations to focus more in learning the generic task-common features. In the later stage, the semantic guidance will be strengthened to learn the task-specific features. Specifically, a dynamic sequential module (DSM) is proposed to explicitly model the sequential relationship and enable multi-stage training. Moreover, to address the side effect of DSM, a new loss regularization is proposed. The extensive experiments on these two clinical datasets show the superiority of the proposed method over the baselines, and demonstrate the effectiveness of the proposed task-coupling elastic mechanism.

Zero-shot learning uses semantic attributes to connect the search space of unseen objects. In recent years, although the deep convolutional network brings powerful visual modeling capabilities to the ZSL task, its visual features have severe pattern inertia and lack of representation of semantic relationships, which leads to severe bias and ambiguity. In response to this, we propose the Graph-based Visual-Semantic Entanglement Network to conduct graph modeling of visual features, which is mapped to semantic attributes by using a knowledge graph, it contains several novel designs: 1. it establishes a multi-path entangled network with the convolutional neural network (CNN) and the graph convolutional network (GCN), which input the visual features from CNN to GCN to model the implicit semantic relations, then GCN feedback the graph modeled information to CNN features; 2. it uses attribute word vectors as the target for the graph semantic modeling of GCN, which forms a self-consistent regression for graph modeling and supervise GCN to learn more personalized attribute relations; 3. it fuses and supplements the hierarchical visual-semantic features refined by graph modeling into visual embedding. Our method outperforms state-of-the-art approaches on multiple representative ZSL datasets: AwA2, CUB, and SUN by promoting the semantic linkage modelling of visual features.

Neural Transducer (e.g., RNN-T) has been widely used in automatic speech recognition (ASR) due to its capabilities of efficiently modeling monotonic alignments between input and output sequences and naturally supporting streaming inputs. Considering that monotonic alignments are also critical to text to speech (TTS) synthesis and streaming TTS is also an important application scenario, in this work, we explore the possibility of applying Transducer to TTS and more. However, it is challenging because it is difficult to trade off the emission (continuous mel-spectrogram prediction) probability and transition (ASR Transducer predicts blank token to indicate transition to next input) probability when calculating the output probability lattice in Transducer, and it is not easy to learn the alignments between text and speech through the output probability lattice. We propose SpeechTransducer (Speech-T for short), a Transformer based Transducer model that 1) uses a new forward algorithm to separate the transition prediction from the continuous mel-spectrogram prediction when calculating the output probability lattice, and uses a diagonal constraint in the probability lattice to help the alignment learning; 2) supports both full-sentence or streaming TTS by adjusting the look-ahead context; and 3) further supports both TTS and ASR together for the first time, which enjoys several advantages including fewer parameters as well as streaming synthesis and recognition in a single model. Experiments on LJSpeech datasets demonstrate that Speech-T 1) is more robust than the attention based autoregressive TTS model due to its inherent monotonic alignments between text and speech; 2) naturally supports streaming TTS with good voice quality; and 3) enjoys the benefit of joint modeling TTS and ASR in a single network.

Online healthcare consultation offers people a convenient way to consult doctors. In this paper, we aim at building a generative dialog system for Chinese healthcare consultation. As the original Seq2seq architecture tends to suffer the issue of generating low-quality responses, the multi-source Seq2seq architecture generating more informative responses is much more preferred in this task. The multi-source Seq2seq architecture takes advantage of retrieval techniques to obtain responses from the database, and then takes these responses alongside the user-issued question as input. However, some of the retrieved responses might be not much related to the user-issued question, resulting in the generation of unsatisfying responses that are not correct in diagnosis or instead provide inappropriate advice on prevention or treatment. Therefore, this paper proposes multi-source Seq2seq guided by knowledge (MSSGK) to handle this problem. MSSGK differs from the multi-source Seq2seq architecture in that domain knowledge, including disease labels and topic labels about prevention and treatment, is introduced into the response generation via a multi-task learning framework. To better exploit the domain knowledge, we propose three attention mechanisms to provide more appropriate guidance for response generation. Experimental results on a dataset of real-world healthcare consultation show the effectiveness of the proposed method.

The tongue image provides important physical information of humans. It is of great importance for diagnoses and treatments in clinical medicine. Herbal prescriptions are simple, noninvasive, and have low side effects. Thus, they are widely applied in China. Studies on the automatic construction technology of herbal prescriptions based on tongue images have great significance for deep learning to explore the relevance of tongue images for herbal prescriptions, it can be applied to healthcare services in mobile medical systems. In order to adapt to the tongue image in a variety of photographic environments and construct herbal prescriptions, a neural network framework for prescription construction is designed. It includes single/double convolution channels and fully connected layers. Furthermore, it proposes the auxiliary therapy topic loss mechanism to model the therapy of Chinese doctors and alleviate the interference of sparse output labels on the diversity of results. The experiment use the real-world tongue images and the corresponding prescriptions and the results can generate prescriptions that are close to the real samples, which verifies the feasibility of the proposed method for the automatic construction of herbal prescriptions from tongue images. Also, it provides a reference for automatic herbal prescription construction from more physical information.

Food recognition plays a much critical role in various health-care applications. However, it poses many challenges to current approaches due to the diverse appearances of food dishes and the non-uniform composition of ingredients for the foods in the same category. Current methods primarily focus on the appearance of foods without considering their semantic information, easily finding the wrong attention areas of food images. Second, these methods lack the dynamic weighting of multiple semantic features in the modeling process. Thus this paper proposes a novel Multi-View Attention Network within the multi-task learning framework that incorporates multiple semantic features into the food recognition task from both ingredient recognition and recipe modeling. It also utilizes the multi-view attention mechanism to automatically adjust the weights of different semantic features and enables different tasks to interact with each other so as to obtain a more comprehensive feature representation. The experiments conducted on both ChineseFoodNet and VIREO Food-172 benchmark databases validate the proposed method with the obvious improvement of the performance and the lower parameter size.

Automatic real-time emotion recognition based on multi-channel EEG signals is a significant and challenging task in neurology and psychiatry. In recent years, deep learning has been used in EEG emotion recognition. However, many existing deep learning based methods still require complex pre-processing or additional feature extraction, which make it difficult to achieve real-time emotion recognition. In this paper, an end-to-end model named Temporal Convolutional Broad Learning System (TCBLS) was designed for multi-channel EEG based emotion recognition. The TCBLS takes one-dimensional EEG signals as input, then extracts emotion-related features of EEG automatically. In this model, the Temporal Convolutional Network (TCN) is designed to extract EEG temporal features and deep abstract features simultaneously, then Broad Learning System (BLS) is used to map the features to a more discriminative space and further enhance the features. We evaluated our method on DEAP database, performing 10-fold cross-validation on each subject to obtain the classification accuracy. Experimental results indicate that the performance of TCBLS is better than other comparison methods, and the mean accuracy of TCBLS is 99.5755% and 99.5781% on valence and arousal classification task respectively. The results demonstrate the effectiveness and robustness of TCBLS in EEG emotion recognition.

Facial expression recognition has been widely used to solve the problems such as lie detection and human-machine interaction. However, due to the difficulties to control the application environments, current methods have the lower recognition accuracy in practice. This paper proposes a new method for facial expression recognition by considering several aspects. First, human beings are easy to recognize some expressions, while difficult to recognize others. Inspired by this intuition, a new loss function is proposed to enlarge the distances between samples from easily confused categories. Second, human learning is divided into many stages, and the learning objective of each stage is different. Thus, dynamic objectives learning is proposed, where each objective at different stage is defined by the corresponding loss function. In order to better realize the above ideas, a new deep neural network for facial expression recognition is proposed, which integrates the covariance pooling layer and residual network units into the deep convolution neural network so as to better perform dynamic objectives learning. The experimental results on the standard databases verify the effectiveness and the superior performance of our methods.

The behavior of the current emotion classification model to recognize all test samples using the same method contradicts the cognition of human beings in the real world, who dynamically change the methods they use based on current test samples. To address this contradiction, this study proposes an individualized emotion recognition method based on context awareness. For a given test sample, a classifier that was deemed the most suitable for the current test sample was first selected from a set of candidate classifiers and then used to realize the individualized emotion recognition. The Bayesian learning method was applied to select the optimal classifier and then evaluate each candidate classifier from the global perspective to guarantee the optimality of each candidate classifier. The results of the study validated the effectiveness of the proposed method.

Recently, hashing methods have been widely used in large-scale image retrieval. However, most existing supervised hashing methods do not consider the hierarchical relation of labels,which means that they ignored the rich semantic information stored in the hierarchy. Moreover, most of previous works treat each bit in a hash code equally, which does not meet the scenario of hierarchical labeled data. To tackle the aforementioned problems, in this paper, we propose a novel deep hashing method, called supervised hierarchical deep hashing (SHDH), to perform hash code learning for hierarchical labeled data. Specifically, we define a novel similarity formula for hierarchical labeled data by weighting each level, and design a deep neural network to obtain a hash code for each data point. Extensive experiments on two real-world public datasets show that the proposed method outperforms the state-of-the-art baselines in the image retrieval task.

To compare the similarity of probability distributions, the information-theoretically motivated metrics like Kullback-Leibler divergence (KL) and Jensen-Shannon divergence (JSD) are often more reasonable compared with metrics for vectors like Euclidean and angular distance. However, existing locality-sensitive hashing (LSH) algorithms cannot support the information-theoretically motivated metrics for probability distributions. In this paper, we first introduce a new approximation formula for S2JSD-distance, and then propose a novel LSH scheme adapted to S2JSD-distance for approximate nearest neighbors search in high-dimensional probability distributions. We define the specific hashing functions, and prove their local-sensitivity. Furthermore, extensive empirical evaluations well illustrate the effectiveness of the proposed hashing schema on six public image datasets and two text datasets, in terms of mean Average Precision, Precision@N and Precision-Recall curve.

Facial expressions recognition (FER) plays a much important role in various applications from human-computer interfaces to psychological tests. However, most methods are confronted with the quality of the face images, vanishing gradients problem, over-trained problem, difference of face images such as in age and ethnicity, mulitple parameters required tuning, and dubious class labels in the training data. These negative factors largely hurt the recognition performance. To alleviate these problems, this paper proposes an new approach named ensemble convolutional echo state network, which takes Echo State Network (ESN) as the base classifier for ensemble and Convolutional Network (CN) to transform the input face image for further feeding to ESN, where the random parameters and architectures are assigned to ensure the diversity of the ensemble and to avoid computing stochastic gradient. Based on the rich dynamics of ESN and rich variations of input face image finished by CN, the proposed approach has the great ability to deal with the real facial expression recognition and to be scaled to the larger training data. It has also only one parameter to be adjusted. Conducted experiments show that the method achieves significant improvement over current methods on person-independent facial expression recognition.

Precise cancer classification is essential to the successful diagnosis and treatment of cancers. Although semisupervised dimensionality reduction approaches perform very well on clean datasets, the topology of the neighborhood constructed with most existing approaches is unstable in the presence of high-dimensional data with noise. In order to solve this problem, a novel local and global preserving semisupervised dimensionality reduction based on random subspace algorithm marked as RSLGSSDR, which utilizes random subspace for semisupervised dimensionality reduction, is proposed. The algorithm first designs multiple diverse graphs on different random subspace of datasets and then fuses these graphs into a mixture graph on which dimensionality reduction is performed. As themixture graph is constructed in lower dimensionality, it can ease the issues on graph construction on highdimensional samples such that it can hold complicated geometric distribution of datasets as the diversity of random subspaces. Experimental results on public gene expression datasets demonstrate that the proposed RSLGSSDR not only has superior recognition performance to competitive methods, but also is robust against a wide range of values of input parameters.

A new learning algorithm for categorical data, named CRN (Classification by Rule-based Neighbors) is proposed in this paper. CRN is a nonmetric and parameter-free classifier, and can be regarded as a hybrid of rule induction and instance-based learning. Based on a new measure of attributes quality and the separate-and-conquer strategy, CRN learns a collection of feature sets such that for each pair of instances belonging to different classes, there is a feature set on which the two instances disagree. For an unlabeled instance I and a labeled instance I', I' is a neighbor of I if and only if they agree on all attributes of a feature set. Then, CRN classifies an unlabeled instance I based on I's neighbors on those learned feature sets. To validate the performance of CRN, CRN is compared with six state-of-the-art classifiers on twenty-four datasets. Experimental results demonstrate that although the underlying idea of CRN is simple, the predictive accuracy of CRN is comparable to or better than that of the state-of-the-art classifiers on most datasets.

Locally linear embedding heavily depends on whether the neighborhood graph represents the underlying geometry structure of the data manifolds. Inspired from the cognitive law, the relative transformation(RT) and kernel relative transformation (KRT) are proposed. They can improve the distinction between data points and inhibit the impact of noise and sparsity of data, which can be then applied to construct the neighborhood graph so as to reduce the short circuit edges, while the embedding is still performed in the original space. Subsequently, another enhanced Hessian Locally Linear Embedding approach is developed with significantly increased performance. The conducted experiments on challenging benchmark data sets validate the proposed approaches.

Text categorization has become the key technology in organizing and processing the large amount of text information. It normally involves an extremely high dimensional space, which makes most existing approaches generate highly biased estimates so as to reduce the classification accuracy. These approaches do not consider that the text documents may be intrinsically located on the low-dimensional manifold. This paper presents an approach that performs text categorization on texts manifold with respect to the intrinsic global manifold structure, such as by geodesic distance to measure the distance between two texts. This approach has been applied to improve the KNN for text categorization. This is empirically validated by the conducted experiments.

Hessian locally linear embedding (HLLE) has good representational capacity and high computational efficiency, but it still fails to nicely deal with the sparsely sampled or noise contaminated datasets, where the local neighborhood structure is critically distorted. To solve this problem, this paper proposes a new approach that takes the general conceptual framework of HLLE so as to guarantee its correctness in the setting of local isometry, and then employs the geodesic distance instead of Euclidean distance to determine the local neighborhood so as to give the global representation to the local data. This approach can be regarded as the integration of both local approaches and global approaches, so that it have the better performance and stability. The conducted experiments on both synthetic and real datasets have validated the proposed approach.

Locally linear embedding approach (LLE) is one of most efficient nonlinear dimensionality reduction approaches with good representational capacity for a broader range of manifolds and high computational efficiency. However, LLE and its variants fail to nicely deal with sparsely sampled or noise contaminated datasets,where the local neighborhood structure is critically distorted. To solve this problem, this paper utilizes the clustering approaches to partition the input data into clusters and then rescale the distance between any points based on the clustering structure so as to make data points from different clusters separated more easily. This rescaled distance matrix is then provided to improve LLE so as to achieve the better performance. Unlike the supervised approaches, this approach does not take the labelled dataset as prerequisite, so that it is unsupervised. This makes it applicable to broader range of domains. The conducted experiments by classification on benchmark datasets have validated the proposed approach.

The paper outlines a self-optimizing approach for knowledge acquisition with adaptively incremental sampling, which fused the self-optimizing approach for knowledge acquisition and sampling approaches in order to improve the efficiency of knowledge acquisition effectively. The proposed sampling approach enabled us to dynamically and adaptively adjust the sample size according to the data mining algorithm's performance on the training samples so as to utilize the sample size as small as possible without reducing the accuracy of the knowledge model. Finally, the self-optimizing approach for knowledge acquisition with adaptively incremental sampling was applied to rule generation from the diagnostic decision table for rheumatoid arthritis in Chinese medical science. Experimentation results showed that the approach was much better than other algorithms both in efficiency and in accuracy.

This paper aims to provide a complete theoretical analysis for the creative methods. First, the feature set is used to express the creative solution, and some of the basic operators induced from different creative methods are defined. Next, the paper employs the information entropy theory to prove the usability of the creative method to produce creative solutions, and presents a detailed proof of the creating ability of the methods in an evolving framework. The prime contribution here is that the paper provides not only the theoretical foundation for the creative methods, but also the way and theoretical foundation for implementation of the system of creative design. Finally, an empirical study demonstrates that solutions, which meet the proposed theory, tend to score high on creativity evaluation by field experts.

Attribute reduction and rule generation (attribute value reduction) are two of the main processes of knowledge acquisition. A self-optimizing approach based on a difference comparison table for knowledge acquisition for these processes is proposed. For the attribute reduction process, conventional logic computation was replaced by matrix computation with some added concepts from evolutionary computation and used to construct the self-adaptive optimizing algorithm. In addition, some sub-algorithms and proofs are presented in detail. For the rule generation process, a value orderly reduction algorithm, which simplifies the complexity of rule knowledge, is presented. The approach provides an effective and efficient method for knowledge acquisition, which is supported by the experimentation.

Automation of creative thinking is very important as well as very difficult. Research shows that creative thinking is a very complicated dynamical system. We first prove that the variation of any feature of this system is able to reflect the dynamic information of the whole system. Based on this, we give a method of how to measure the reminded objects so that the measure sequence can be constructed which is proved to be chaotic. It follows that we present a creative reminding method according to chaos theory. This method is built on the prediction by a Lyapunov exponent calculated on the thinking history, which consists of the exponent calculation, exponent reminding mode, etc. We present a comprehensive reminding algorithm consisting of the usual and fantastic parts. When the relevance of the target to the given base is bigger than the given threshold the target is to be reminded, or else a simulated annealing-based algorithm is applied to make a decision. When the system fails continuously many times to produce reminders, the system will try to employ the exponent prediction to produce the fantastic reminder which often results in inspiration. The proposed method can be used to find the conditions leading to the inspiration so as to provide the means for the system to yield creative thinking. Now the proposed method has been integrated into the prototype system of creative design. The results of the experiment show its advantages in some aspects.

In this paper an evolution model for creative design involving a mathematical model, evolution framework and implementation algorithms etc., is presented to effectively support the automation of creative design. The model takes the creative design process as an evolving process of a product prototype to the optimal scheme in accordance with a kind of measurement. This evolving process is built on the open genetic algorithm so as not to be confined by the initial population, which can get new individuals beyond the initial population by using creative reminding algorithms. At the same time, the model not only defines new operators based on the semantics of invention such as crossover, mutation, fitness function etc., but also provides an interface for integrating some other new methods of creative design as well as creative thinking patterns into the system. This model also supports a human-machine interface for receiving the guide from the design engineer online. Results of experiments with our implemented system, named AIE1.0, show that the model is promising.