Data integrity verification in geographically distributed edge systems remains a critical unsolved challenge. While centralized verification introduces bottlenecks and single points of failure, existing decentralized alternatives suffer from inefficiency due to their lack of batch verification capabilities. This limitation leads to prohibitive communication and computational overheads that scale poorly as data volume grows. This paper introduces EdgeBatch, the first decentralized protocol designed for efficient batch integrity verification, reducing communication rounds from$\mathcal {O}(n)$to$\mathcal {O}(1)$a small, constant number. At its core is a reputation-aware Combination Selection Algorithm (CSA), a polynomial-time heuristic that identifies near-optimal peer server combinations, balancing verifier group size against servers' historical trustworthiness through intelligent pruning strategies. This process is orchestrated through distributed ledger technology and smart contracts, ensuring a secure, transparent, and trustless verification environment. The protocol's design is underpinned by rigorous theoretical analysis, including formal proofs of security and correctness, and a probabilistic model for optimizing key system parameters. Extensive simulations show that EdgeBatch drastically outperforms state-of-the-art methods; it improves computational efficiency by an average of 518.60× over EdgeWatch and 1030.93× over CooperEDI, while also reducing communication overhead by 296.68× and 62.66×, respectively. A concluding ablation study confirms the vital role of our reputation mechanism, demonstrating it reduces the required verification rounds by 73% and is the key to the protocol's efficiency.

Due to its computational efficiency and low storage requirement, cross-modal hashing (CMH) gains a lot of attention. However, there are still three issues that affect its performance: 1) most existing methods focus on learning the shared semantics between different modalities with the modality-specific semantics ignored; 2) noisy labels may further exacerbate the semantic differences of different modalities during learning; and 3) most existing methods overlook the complementarity between modality-specific labels and semantic features. To address these issues, this work develops a novel CMH method called Dual-Semantic Enhancement Cross-Modal Hashing with Noisy Labels (DSENL). DSENL consists of three parts: (a) Modality-Specific Label Recovery (MSLR) that obtains modality-specific clean labels by applying matrix decomposition with low-rank and sparse constraints to the observed labels; (b) Semantic Preservation under Label Guidance (SPLG) that enhances the quality of recovered labels by using an$l_{2,1}$norm and maintains semantic consistency across modalities by reducing discrepancies among modality-specific labels; and (c) Dual-Semantic Enhancement Learning (DSEL) that integrates both label and sample semantics from modality-specific to enhance the discriminative capability of hash codes. By DSENL, the discriminability of the learned hash codes is improved. Experimental results on four benchmark datasets demonstrate the effectiveness of DSENL. The source code is available athttps://github.com/niuniubit/DSENL.git.

Domain adaptation (DA) is designed to tackle the problem of label scarcity in the target domain by transferring knowledge to it. However, there are two critical challenges in DA : 1) insufficient discriminative power of learned features, and 2) inadequate exploration of inter-sample relationships. This study proposes a novel framework, Joint Consistency-Driven Dual-Domain Discriminative Learning (JCD3L) to overcome these limitations. This framework encompasses two components: Inter-domain Collaborative Feature Enhancement (ID-CFE) and Joint Semantic-Spatial Consistency Constraint (JSSCC). Firstly, ID-CFE applies angular margin (AM) loss to the source domain while imposing entropy regularization on the target domain, establishing a dual-domain discriminative enhancement mechanism for feature representations. Additionally, a novel consistency regularization, JSSCC, is proposed to thoroughly explore the interrelationships among samples. This regularization leverages the label-semantics and feature-semantics to refine the alignment process. To verify the effectiveness of our work, comprehensive experiments are conducted across three widely used benchmarks and the results demonstrate considerable improvements.

With the rapid growth of multi-modal data, Cross-Modal Hashing (CMH) is widely applied due to its outstanding performance in both search and storage. Nevertheless, there are two issues to be further addressed: (1) most existing methods neglect dynamic learning of the importance of different labels; and (2) many methods fail to purify the consistency of data extracted from different feature spaces. For this purpose, we propose a method called Dynamic Label Correlations and Dual-Semantic Enhancement Learning for Cross-Modal Retrieval (DLCDE) in this study. This method is formed of two parts: Label Semantic Enhancement with Dynamic Label Reconstruction (LSEDLR) and Sample Semantic Enhancement with Consistency Purification and Structure Maintenance (SECPSM). The former first utilizes label-wise self-expression to dynamically explore the latent correlations between different labels and then employs a graph-based manifold regularizer to explore the structural relationships in the transformed label space to enhance label semantics, the latter leverages Hadamard-Product-based Matrix Factorization to enhance the common relationships between samples, thereby enhancing the sample semantics of the latent shared space. Moreover, dual-semantic enhancement learning is achieved by integrating enhanced label semantics and sample semantics in Distance-Distance Difference Minimization (DDDM). Numerous experiments on four benchmark datasets reveal that DLCDE surpasses a number of state-of-the-art CMH methods . The source code for DLCDE is publicly available at https://github.com/Fizzyf/DLCDE .

Domain Adaptation (DA) is used to reduce cross-domain differences between the labeled source and unlabeled target domains. As the existing semantic-based DA approaches mainly focus on extracting consistent knowledge under semantic guidance, they may fail in acquiring (a) personalized knowledge between intra-class samples, and (b) local knowledge of neighbor samples from different categories. Hence, a multi-semantic-granularity and target-sample oriented approach, called Adaptive Graph Learning with Semantic Promotability (AGLSP), is proposed, which consists of three parts: (a) Adaptive Graph Embedding with Semantic Guidance (AGE-SG) that adaptively estimates the promotability of target samples and learns variant semantic and geometrical components from the source and those semantically promotable target samples; (b) Semantically Promotable Sample Enhancement (SPSE) that further increases the discriminability and adaptability of tag granularity by mining the features of intra-class source and semantically promotable target samples with multi-granularities; and (c) Adaptive Graph Learning with Implicit Semantic Preservation (AGL-ISP) that forms the tag granularity by extracting commonalities between the source and those semantically non-promotable target samples. As AGLSP learns more semantics from the two domains, more cross-domain knowledge is transferred. Mathematical proofs and extensive experiments on seven datasets demonstrate the performance of AGLSP.

Ensuring data integrity in edge computing environments presents significant challenges, primarily due to the distributed architecture of edge servers and the inherent risk of data corruption. Traditional edge data integrity (EDI) verification methods predominantly rely on sampling techniques, provide only probabilistic integrity assurances and often struggle with scalability and efficient corruption localization. To overcome these limitations, we introduce the deterministic integrity assurance and rapid corruption localization EDI (DL-EDI) verification scheme, a novel approach that combines extended Merkle grid (EM-Grid) with Boneh–Lynn–Shacham (BLS) signatures. DL-EDI leverages EM-Grid for comprehensive data integrity verification, ensuring deterministic integrity validation across all data blocks while facilitating rapid block corruption localization. Additionally, we incorporate a hierarchical signature aggregation method using BLS signatures to optimize verification efficiency and minimize communication and computational overhead. A thorough performance analysis of DL-EDI is conducted, evaluating its verification accuracy, communication and computational efficiency, and resilience against various security threats. Comparative experimental evaluations of DL-EDI against four established EDI schemes highlight its superior effectiveness and efficiency in addressing the challenges of EDI.

Due to the large-scale and multiscale natures of social media data, sparse representation (SR) learning methods are widely followed. However, there are three problems associated with the existing SR methods: 1) they neglect the fact that the semantic features of data may change during iterative learning, which leads to weak semantic learning; 2) they often assume that the data are linearly separable, while the data might be nonlinear in many real-world applications; and 3) they cannot ensure the low-rank and discriminative properties of the data at the same time and might neglect the global properties of the data, leading to suboptimal solutions. To solve these problems, we propose a novel method, named kernel-based SR learning with global and local low-rank label (KSR-GL3) constraint, which strengthens the semantic information and ensures the semantic features invariant during learning. First, we map the data into a high-dimensional feature space to learn the linear representation of samples. Second, global and local low-rank label (GL3) constraint is used to ensure the semantic invariance, low-rankness, and discrimination of features during learning. Third, an$\ell _{2,1}$is imposed to explore the sparseness of the subspace. Mathematical analyses show that GL3 can retain the intrinsic properties of data during learning. By combining the above three components, a generalized power iteration (GPI) approach is applied to build the model and deal with the tricky optimization problem. By KSR-GL 3, a sparse, low-rank, and discriminative subspace is produced from the high-dimensional and orthogonal representation of the data under the guidance of semantics, while the intrinsic properties of data are preserved. Extensive experiments on six datasets compared with five advanced algorithms demonstrate its promising prospects.

As social media faces with large amounts of data and multimodal properties, cross-modal hashing (CMH) retrieval gains extensive applications with its high efficiency and low storage consumption. However, there are two issues that hinder the performance of the existing semantics-learning-based CMH methods: 1) there exist some nonlinear relationships, noises, and outliers in the data, which may degrade the learning effectiveness of a model; and 2) the complementary relationships between the label semantics and sample semantics may be inadequately explored. To address the above two problems, a method called robust asymmetric cross-modal hashing retrieval with dual semantic enhancement (RADSE) is proposed. RADSE consists of three parts: 1) cross-modal data alignment (CDA) that applies kernel mapping and establishes a unified linear representation in the neighborhood to capture the nonlinear relationships between cross-modal data; 2) relaxed label semantic learning for robustness (RLSLR) that uses a relaxation strategy to expand label distinctiveness, and leverages$\ell_{2,1}$norm to enhance the robustness of the model against noise and outliers; and 3) dual semantic enhancement learning (DSEL) that learns more interrelationships between samples under the label semantic guidance to ensure the mutual enhancement of semantic information. Extensive experiments and analyses on three popular datasets demonstrate that RADSE outperforms the most existing methods in terms of mean average precision (MAP), precision recall (P–R) curves, and top-N precision curves. In the comparisons of MAP, RADSE improves by an average of 2%–3% in two retrieval tasks.

Due to high computational efficiency and low storage cost, cross-modal hashing retrieval attracts much attention. However, as heterogeneous data from different modalities often have distinct physical meanings and underlying structures, learning encoding with equal length for different modalities may result in an insurmountable semantic gap. In addition, there are still some issues, e.g., how to combine label and sample information to learn hash codes effectively, how to reduce the time consumption caused by computing n × n similarity matrix, and how to effectively solve the complex discrete optimization problem. To overcome the above challenges, this study propose a novel model called Scalable Discrete and Asymmetric Unequal Length Hashing (SDAULH). First, SDAULH constructs a novel hash model that utilizes unequal length encoding schemes to narrow the semantic gap between heterogeneous modalities. Second, SDAULH develops a dual semantic embedding learning scheme, which combines pairwise similarity between label and sample data to generate a more discriminative hash code. Third, SDAULH associates with both hash codes and label information by an asymmetric relaxation strategy. Furthermore, SDAULH solves directly the discrete optimization problem by generating discrete hash codes. Experimental results on four benchmark datasets demonstrate the promising performance of SDAULH.

To accurately retrieve similar objects from different domains, domain adaptive retrieval method is applied to cope with the domain shift problem in information retrieval. However, existing methods still have two problems: a) they fail to filter out low-confidence samples, leading to error accumulation; and b) they ignore the negative effect of domain discrepancy. To address these two issues, we propose an efficient method called Dynamic Confidence Sampling and Label Semantic Guidance Learning (DCS-LSG). First, Dynamic Confidence Sampling (DCS) is employed to dynamically select high-confidence samples from the target domain so as to improve the effectiveness of learning. Second, Label Semantic Guidance (LSG) learning is presented to enhance the label semantics of features during domain adaptive retrieval. In addition, we introduce a Dual-Projection Relaxation (DPR) strategy to learn more effective features on two specific projection spaces. At last, a two-step hashing strategy is used to generate high-quality hash codes. Experiments on multiple cross-domain retrieval datasets demonstrate that the proposed DCS-LSG can achieve a significant performance improvement.

Mobile edge computing allows for high-performance and low-latency applications by delegating computation and data processing tasks to edge servers. However, ensuring the integrity of cached data on these servers can be challenging due to their limited resources. Current designs often use a per-edge multi-round approach, which necessitates multiple communication rounds between each edge server and the application vendor (AppVend). This approach results in high communication and computational costs, as well as the stragglers effect during batch verification. To address these inefficiencies, we propose a Hierarchical Signature Aggregation for Edge Data Integrity (HSA-EDI) verification design. Our design adopts a novel per-edge one-round approach, which significantly reduce the number of communication rounds to one for each edge server, while mitigating the impact of stragglers. Furthermore, it remarkably reduces computational costs through a hierarchical aggregation mechanism. This mechanism supports intra-edge signature aggregation at the edge server level, followed by inter-edge aggregation at the AppVend, which enhances overall efficiency. We then conduct a theoretical analysis of HSA-EDI’s correctness, security, and communication, computation, and storage efficiency. Experimental results validate its superior performance over state-of-the-art designs.

Hashing has attracted widespread attention in the field of supervised cross-modal retrieval due to its advantages in search and storage. However, there are still some issues to be addressed, e.g.: 1) how to effectively combine sample and label semantics to learn hash codes; 2) how to reduce high computational requirements brought by computing a pairwise similarity matrix; and 3) how to effectively solve discrete optimization problems. To cope with them, a fast asymmetric and discrete cross-modal hashing (FADCH) method is proposed in this article. First, matrix factorization is leveraged to collaboratively construct a common semantic subspace between different modalities. Second, semantic consistency is preserved by aligning the common semantic subspace with the semantic representation constructed from labels, which effectively exploits the semantic complementarity of labels and samples. Third, we embed labels into hash codes and keep the correlation between different modal samples by using a pairwise similarity matrix. Fourth, we use an asymmetric strategy with relaxation to associate hash codes with semantic representation, which not only avoids the difficulty of symmetric frame optimization but also embeds more semantic information into the Hamming space. In addition, a strongly orthogonal constraint is introduced to optimize the hash codes. Finally, an effective optimization algorithm is developed to directly generate discrete hash codes while reducing the complexity from$O(n^{2})$to$O(n)$. The experimental results on three benchmark datasets illustrate the superiority of the FADCH method.

In general, existing cross-domain recognition methods mainly focus on changing the feature representation of data or modifying the classifier parameter and their efficiencies are indicated by the better performance. However, most existing methods do not simultaneously integrate them into a unified optimization objective for further improving the learning efficiency. In this article, we propose a novel cross-domain recognition algorithm framework by integrating both of them. Specifically, we reduce the discrepancies in both the conditional distribution and marginal distribution between different domains in order to learn a new feature representation which pulls the data from different domains closer on the whole. However, the data from different domains but the same class cannot interlace together enough and thus it is not reasonable to mix them for training a single classifier. To this end, we further propose to learn double classifiers on the respective domain and require that they dynamically approximate to each other during learning. This guarantees that we finally learn a suitable classifier from the double classifiers by using the strategy of classifier fusion. The experiments show that the proposed method outperforms over the state-of-the-art methods.

Cross-modal retrieval has attracted considerable attention for searching in large-scale multimedia databases because of its efficiency and effectiveness. As a powerful tool of data analysis, matrix factorization is commonly used to learn hash codes for cross-modal retrieval, but there are still many shortcomings. First, most of these methods only focus on preserving locality of data but they ignore other factors such as preserving reconstruction residual of data during matrix factorization. Second, the energy loss of data is not considered when the data of cross-modal are projected into a common semantic space. Third, the data of cross-modal are directly projected into a unified semantic space which is not reasonable since the data from different modalities have different properties. This article proposes a novel method called average approximate hashing (AAH) to address these problems by: 1) integrating the locality and residual preservation into a graph embedding framework by using the label information; 2) projecting data from different modalities into different semantic spaces and then making the two spaces approximate to each other so that a unified hash code can be obtained; and 3) introducing a principal component analysis (PCA)-like projection matrix into the graph embedding framework to guarantee that the projected data can preserve the main energy of data. AAH obtains the final hash codes by using an average approximate strategy, that is, using the mean of projected data of different modalities as the hash codes. Experiments on standard databases show that the proposed AAH outperforms several state-of-the-art cross-modal hashing methods.

Existing visual-based fatigue detection methods usually monitor drivers’ fatigue by capturing their facial features, including eyelid movements, yawn frequency and head pose. However, these approaches typically do not take drivers’ biological signals into consideration. An accurate model for fatigue detection requires combining both facial behavior and biological data. This paper proposes a novel non-intrusive method for driver multimodal fusion fatigue detection by extracting eyelid features and heart rate signals from the RGB video. The multimodal feature fusion method could significantly increase the accuracy of fatigue detection. Specifically, we established two fatigue detection models based on heart rate and the PERCLOS value respectively with one-dimensional Convolutional Neural Network (1D CNN), where the PERCLOS refers to the percentage of eyelid closure over the pupil. Finally, the outputs of the two models are weighted to achieve the multimodal fusion fatigue detection. Simulation results show that our method yield better performance than traditional methods.

Sentiment classification has been attracting increasing attention with the growth of textual data created on the Internet. Text review data covers a wide range of field, and sentiment classification has been widely known as a highly domain-dependent problem. Unfortunately, the existing methods have achieved good results in the domain with a large number of labeled training data. Some researchers apply classifiers learned from source domain to target domain through transfer learning, which still requires the target domain to have enough unlabeled data to learn the similarity between the domains. In this paper, we propose a collaborative domain-specific and global multi-domain sentiment classification approaches with logistic regression. We train a domain-specific sentiment classifier for each source domain, reconstruct the source domain datasets, and train the global sentiment classifiers. Domain-specific sentiment classifier captures domain-specific sentiment features, and global sentiment classifier captures general sentiment knowledge. Finally, taking the output of the first layer as the input of the second layer, a two-level cross-domain sentiment classification model is constructed by logistic regression. Experimental results on benchmark datasets show that the proposed approach can effectively improve the performance of multi-domain sentiment classification and significantly outperform baseline methods.

In the field of host-based intrusion detection systems(HIDS), existing anomaly detection algorithms paid much attention to extracting system call features, such as N-gram, frequency-based and neural networks, whereas few literatures introduce effective methods of modeling system call sequences with semantic features. This paper proposes a new model to represent system call sequences with novel applications of embedding techniques. The method converts system calls into embedding vectors as inputs for anomaly detector, mainly includes two parts: 1) construct embedding vectors for all system calls; 2) model the sequences with system call embedding and weighting. This sequence representation model is intuitive and effective on ADFA-LD dataset. As is illustrated with our experiment, the FPR can be significantly reduced to 0.53%, while the TPR still reaches 91.7%, even by a simple 1-NN classifier.

Unsupervised Domain Adaptation (UDA) has become a basic technology for cross-domain recognition and has received extensive attention in recent years. UDA aims to obtain a classifier for the target domain by learning source instances with different data distributions. However, traditional domain adaptation algorithms cannot effectively explore the manifold structure of data while reducing the distribution differences between domains. To address this problem, this paper proposes a new UDA framework called Joint Discriminative Distribution Adaptation and Manifold Regularization (DDAMR). DDAMR makes full use of the category information and geometric structure of samples in the Grassmann manifold to learn the domain-invariant classifier. Specifically, DDAMR performs discriminative distribution adaptation during dynamic distribution calibration to enhance the discrimination ability of the feature space. In addition, DDAMR introduces manifold regularization that can maintain the proximity relationship of the samples. It can maximize effectively the consistency between the prediction structure of the domain-invariant classifier f and the inherent manifold structure of the sample. A large number of results from cross-domain experiments have demonstrated the effectiveness of our DDAMR algorithm.

Medical resources of physical examination (P.E.) are often insufficient. The gap between providers and demanders are always existing and becoming more and more sensitive in some densely populated areas. If resources of P.E. departments are regarded as nodes, then each path selected by patients will constitute a small world network. Furthermore, with respect to traditional research concentrating on the feature between nodes in network, adaptive collaboration (AC) is in fact an important method to improve the group performance of the whole system in the small world network. Based on these, this paper deals with this kind of problem with respect to the scenario of P.E., which attempts to help decision makers of the health center to schedule limited resources and improve a patient's satisfaction and the system performance. It firstly abstracts a medical examination by Role-Based Collaboration (RBC) and its general model E-CARGO. The adaptive collaboration model is constructed by system states and optimized via series of group role assignments (GRAs), which is a subtask of RBC, and it can be accomplished by linear programming. All the proposed methods are verified by simulation experiments, and the team performance is improved via adaptation of the assignment strategies, which provides a new insight into the small world study.

Human age estimation from face image has drawn increasing research attention due to its many meaningful applications such as demographics analysis and surveillance monitoring. However, most existing methods directly extract age-specific features for age estimation and ignore age-related gender information. In this paper, we propose a simplified deep learning network for age estimation by simultaneously learning aging and potential gender features. Specifically, we first learn the potential gender information from face images. Then, we employ a two-stream convolutional neural network to simultaneously learn and concatenate the aging and gender latent appearance features. Third, we feed the multi-type features into a compact convolution network, named AgeNetwork, to further learn the age-specific features. Finally, we use a deep regression function to estimate the detailed ages. Extensive experimental results demonstrate the promising effectiveness and efficiency of our proposed method in comparison with state-of-the-arts.

Incomplete multi-view clustering has attracted increasing attentions due to its superiority in partitioning unlabeled multi-view data with missing instances in real application. However, most existing methods cannot fully exploit both the view-specific and cross-view relations among data points and ignore the high-order correlations across all views. To address these issues, we propose a novel Incomplete Multi-view Subspace Clustering with Low-rank Tensor (IMSCLT) method, which could be the first tensor-based incomplete multi-view clustering method to the best of our knowledge. Specifically, the subspace representations with low-rank tensor constraint are employed to exploit both the view-specific and cross-view relations among data points and capture the high-order correlations of multiple views simultaneously. In addition, we devise a novel module which can learn a discriminative similarity graph for multi-view learning task by approximating the inner product of the view-specific and common subspace representations. Augmented Lagrangian alternative direction minimization strategy is adopted to solve the proposed IMSCLT. The experiments on several benchmark datasets demonstrate the effectiveness of IMSCLT.

Cross-modal hashing has attracted much attention for cross-modal retrieval and achieved promising performance due to its powerful capacity. Some existing cross-modal hashing methods construct pairwise similarities to represent the relationship of heterogeneous data, which require much computation time and storage space, making them unscalable for large-scale retrieval tasks. In this paper, we propose a novel supervised Discrete Semantic Embedding Hashing (DSEH) for cross-modal retrieval. Specifically, we first learn the common representation of heterogeneous data by embedding the semantic labels into a collective matrix factorization, such that both intra- and inter-modality similarities can be well captured. Then, we learn the hash codes in the discrete space based on the learned common representation via an orthogonal rotation technique. Moreover, we learn the multi-modal hash functions that can efficiently convert out-of-sample instances into unified hash codes. Extensive experimental results on three widely used benchmark databases demonstrate the superiority of the proposed DSEH compared with previous state-of-the-arts.

Oil resources are becoming increasingly scarce. Pure electric vehicles have huge advantages, in terms of energy efficiency and emission reduction. In cities, the locations of required charging stations and the number of required charging piles are determined according to the traffic flow on a road. Unreasonable allocation not only creates safety problems due to high electrical loads, but also increases the cost of the placements. Such allocations will involve the many-to-many (M2M) assignment in the process, which is necessary to establish an optimal model for distributing. Thus, this paper formalizes the charging pile siting problem (CPSP) via the group multirole assignment (GMRA) model, which is one of the most important methods to deal with the M2M problem. Based on GMRA, this paper proposes a role negotiation method by using a spectral clustering K-Means++ Algorithm based on location. The formalization of GMRA makes it easy to find a solution using the IBM ILOG CPLEX optimization package (CPLEX) via the Integer Programming (IP). All the proposed approaches are verified by simulation experiments, which have been proved to be efficient, feasible and practicable.

Hashing has been widely studied for cross-modal retrieval due to its promising efficiency and effectiveness in massive data analysis. However, most existing supervised hashing has the limitations of inefficiency for very large-scale search and intractable discrete constraint for hash codes learning. In this paper, we propose a new supervised hashing method, namely, Discrete Semantic Matrix Factorization Hashing (DSMFH), for cross-modal retrieval. First, we conduct the matrix factorization via directly utilizing the available label information to obtain a latent representation, so that both the inter-modality and intra-modality similarities are well preserved. Then, we simultaneously learn the discriminative hash codes and corresponding hash functions by deriving the matrix factorization into a discrete optimization. Finally, we adopt an alternatively iterative procedure to efficiently optimize the matrix factorization and discrete learning. Extensive experimental results on three widely used image-tag databases demonstrate the superiority of the DSMFH over state-of-the-art cross-modal hashing methods.

Dictionary learning plays a significant role in the field of machine learning. Existing works mainly focus on learning dictionary from a single domain. In this paper, we propose a novel projective double reconstructions (PDR) based dictionary learning algorithm for cross-domain recognition. Owing the distribution discrepancy between different domains, the label information is hard utilized for improving discriminability of dictionary fully. Thus, we propose a more flexible label consistent term and associate it with each dictionary item, which makes the reconstruction coefficients have more discriminability as much as possible. Due to the intrinsic correlation between cross-domain data, the data should be reconstructed with each other. Based on this consideration, we further propose a projective double reconstructions scheme to guarantee that the learned dictionary has the abilities of data itself reconstruction and data crossreconstruction. This also guarantees that the data from different domains can be boosted mutually for obtaining a good data alignment, making the learned dictionary have more transferability. We integrate the double reconstructions, label consistency constraint and classifier learning into a unified objective and its solution can be obtained by proposed optimization algorithm that is more efficient than the conventional l1 optimization based dictionary learning methods. The experiments show that the proposed PDR not only greatly reduces the time complexity for both training and testing, but also outperforms over the stateof- the-art methods.

Ensemble classifier, b y combining multiple classifiers, can often achieve better performance than single classifiers in intrusion detection. Although some ensemble methods have been used for intrusion detection, most of them directly fuse detection outputs after multiple classifiers a re generated. It potentially reduce the overall performance and flexibility. Aiming at achieving a high-precision intrusion detection model with good generalization performance and robustness, an error-correcting ability based collaborative multi-layer selective classifier ensemble model is proposed in this paper, named ML-SCEM. In the ML-SCEM, a novel multi-layer structure consisting of 5 continuous layers is designed, each layer of which is equivalent to a binary classification. In each layer, an error-correcting based selective classifier ensemble method(SCEM) is used to select the main classifier and error-correcting components from M preselected base classifiers to generate an ensemble classifier suitable for this layer classification category. Furthermore to improve time efficiency a nd detection performance, the original dataset is divided into 3 parts of TCP, UDP and ICMP according to the network protocol, so that the three parts are collaboratively detected. The performance of the proposed ML-SCEM is evaluated and compared on the NSL-KDD dataset. It achieves accuracy of 97.07%, false positive rate of 1.58% and efficiently detects various types of attacks.

Direction information has been intensively investigated for Finger-Knuckle-Print (FKP) recognition. However, most existing direction-based KFP recognition methods are handcrafted, which are heuristic and require too much prior knowledge to engineer them. In this paper, we propose a discriminative direction binary feature learning (DDBFL) method for FKP recognition. We first propose a direction convolution difference vector (DCDV) to better describe the direction information of FKP images. Then, we learn a feature projection to convert the DCDV into binary codes, which are compact for the intra-class samples and more separable for the inter-class samples. Finally, we concatenate the block-wise histograms of the DDBFL codes to form the final descriptor for FKP recognition. Experimental results on the baseline PolyU FKP database demonstrate the competitive performance of the proposed method.

In this paper, a Catboost-based framework is proposed to predict social media popularity. The framework is constituted by two components: feature representation and Catboost training. In the component of feature representation, numerical features are directly used, while categorical features are converted into numerical features by a method of order target statistics in Catboost. Besides, some additional user information is also tracked to enrich the feature space. In the other component, Catboost is adopted as the regression model which is trained by using post-related, user-related and additional user information. Moreover, to make full use of the dataset for model training, a dataset augmentation strategy based on pseudo labels is proposed. This strategy involves in two-stage training. In the first stage, it trains a first-stage model that is used to label the test set as pseudo labeled. In the next stage, a final model is trained based on the new training set that includes original validation set and the pseudo labeled test set. The proposed method achieves the 2nd place in the leader board of the Grand Challenge of Social Media Prediction.

In this paper, we propose a unified model called flexible affinity matrix learning (FAML) for unsupervised and semisupervised classification by exploiting both the relationship among data and the clustering structure simultaneously. To capture the relationship among data, we exploit the self-expressiveness property of data to learn a structured matrix in which the structures are induced by different norms. A rank constraint is imposed on the Laplacian matrix of the desired affinity matrix, so that the connected components of data are exactly equal to the cluster number. Thus, the clustering structure is explicit in the learned affinity matrix. By making the estimated affinity matrix approximate the structured matrix during the learning procedure, FAML allows the affinity matrix itself to be adaptively adjusted such that the learned affinity matrix can well capture both the relationship among data and the clustering structure. Thus, FAML has the potential to perform better than other related methods. We derive optimization algorithms to solve the corresponding problems. Extensive unsupervised and semisupervised classification experiments on both synthetic data and real-world benchmark data sets show that the proposed FAML consistently outperforms the state-of-the-art methods.

Palmprint processes a number of unique features for reliable personal recognition. However, different types of palmprint images contain different dominant features. Instead, only some features of the palmprint are visible in a palmprint image, whereas the other features may not be notable. For example, the low-resolution palmprint image has visible principal lines and wrinkles. By contrast, the high-resolution palmprint image contains clear ridge patterns and minutiae points. In addition, the three dimensional (3-D) palmprint image possesses curvatures of the palmprint surface. So far, there is no work to summarize the feature extraction of different types of palmprint images. In this paper, we have an aim to completely study the feature extraction and recognition of palmprint. We propose to use a unified framework to classify palmprint images into four categories: (1) the contact-based; (2) contactless; (3) high-resolution; and (4) 3-D palmprint images. Then, we analyze the motivations and theories of the representative extraction and matching methods for different types of palmprint images. Finally, we compare and test the state-of-the-art methods via the widely used palmprint databases, and point out some potential directions for future research.

Cybersecurity, especially intrusion detection, is becoming increasingly critical in our daily life. The intrusion detection systems (IDS) have been widely used to prevent disclosure of personal information and detect potentially suspicious attacks. Although many machine learning algorithms have been broadly applied to enhance the performance of IDS, low detection rate and high false alarm rate are still two critical problems. A collaborative and robust intrusion detection model using a novel optimal weight strategy based on Genetic Algorithm (GA) for ensemble classifier is proposed in this paper. Since network data stream can be divided into three categories according to network protocols, detectors are applied in the network protocol separately. All of the detectors can work collaboratively and efficiently. In the proposed model, GA is used to optimize the weight of each base classifier of ensemble classifier. In order to improve features quality, Principal Component Analysis (PCA) is used for dimension reduction and attribute extraction. The NSL-KDD datasets is used to test the effectiveness of the collaborative intrusion detection model. Experimental results show that the proposed model has a higher accuracy and better generalized performance than others in this field.

Role assignment is a critical element in the role-based collaboration process. There are many factors to consider when decision makers undertake this task. Such factors include a decision maker's preferences and the team's performance. This paper proposes a series of methods, relative to these factors, to solve the group role assignment with balance problem through an association with the one clause at a time approach that is a well-accepted and logic-based association rule mining method. The proposed methods are verified by simulation experiments. The experimental results present the practicability of the proposed solutions. Using the proposed methods, decision makers need only to establish coarse-grain preferences. The fine-grain preferences can be mined. Furthermore, a balance is obtained between the fine-grain preferences and the team's performance.

This paper proposes a novel method, called robust latent subspace learning (RLSL), for image classification. We formulate an RLSL problem as a joint optimization problem over both the latent SL and classification model parameter predication, which simultaneously minimizes: 1) the regression loss between the learned data representation and objective outputs and 2) the reconstruction error between the learned data representation and original inputs. The latent subspace can be used as a bridge that is expected to seamlessly connect the origin visual features and their class labels and hence improve the overall prediction performance. RLSL combines feature learning with classification so that the learned data representation in the latent subspace is more discriminative for classification. To learn a robust latent subspace, we use a sparse item to compensate error, which helps suppress the interference of noise via weakening its response during regression. An efficient optimization algorithm is designed to solve the proposed optimization problem. To validate the effectiveness of the proposed RLSL method, we conduct experiments on diverse databases and encouraging recognition results are achieved compared with many state-of-the-arts methods.

Role assignment is a critical element in the role-based collaboration process. There are many different requirements to be considered when undertaking this task. This correspondence paper formalizes the group multirole assignment (GMRA) problem; proves the necessary and sufficient condition for the problem to have a feasible solution, provides an improved IBM ILOG CPLEX optimization package solution, and verifies the proposed solution with experiments. The contributions of this paper include: 1) the formalization of an important engineering problem, i.e., the GMRA problem; 2) a theoretical proof of the necessary and sufficient condition for GMRA to have a feasible solution; and 3) an improved ILOG solution to such a problem.

Nowadays, as the development of mobile communication, it is very important to serve users. Because of the subjectivity of user experience, the data of user experience has deviation. In the paper, a cooperative modeling method based on the improved Support Vector Machine is proposed, which can evaluate the quality of experience by using measurement report. The results of the experiments show that our method is effective to calculate the quality of experience.

Natural disasters can lead to severe losses in human life and property. Because many factors combine in a disaster, such events are difficult to forecast accurately. A cooperative multi-classifier method is proposed in this paper to mine local area meteorological data. The proposed method is verified by the implementation of both base and integration classifiers. Experimental results indicate that our proposed method has higher classification accuracy and faster grouping ability compared with conventional classifiers.

Role-Based Collaboration (RBC) is a computational thinking methodology where roles provide an underlying mechanism to facilitate abstraction, classification, separation of concerns, dynamics, and interactions. From a meta theoretical perspective, the specification of groups, roles and agents is a critical element of RBC. In consideration of the relationships and hierarchies faced by groups, roles and agents, we propose a minimal role playing logic system (MRPL) through substructural logic, which is polynomial in complexity, in support of RBC. From MRPL and RPLs extending from it, there are three levels of application, i.e., the global level governing how people organize agents to form a group; the concatenative level for role assignment with respect to logic and algebra, and the operational level governing properties, relations and structures that should appear in collaborative system design. From MRPL to RPLs, one can extend it to suit other appropriate applications.

The image of a face varies with the illumination, pose, and facial expression, thus we say that a single face image is of high uncertainty for representing the face. In this sense, a face image is just an observation and it should not be considered as the absolutely accurate representation of the face. As more face images from the same person provide more observations of the face, more face images may be useful for reducing the uncertainty of the representation of the face and improving the accuracy of face recognition. However, in a real world face recognition system, a subject usually has only a limited number of available face images and thus there is high uncertainty. In this paper, we attempt to improve the face recognition accuracy by reducing the uncertainty. First, we reduce the uncertainty of the face representation by synthesizing the virtual training samples. Then, we select useful training samples that are similar to the test sample from the set of all the original and synthesized virtual training samples. Moreover, we state a theorem that determines the upper bound of the number of useful training samples. Finally, we devise a representation approach based on the selected useful training samples to perform face recognition. Experimental results on five widely used face databases demonstrate that our proposed approach can not only obtain a high face recognition accuracy, but also has a lower computational complexity than the other state-of-the-art approaches.

We firstly analyze the method for four attack types, including Probing, DoS (Denial of Service), R2L (Remote to Local) and U2R (User to Root). Based on resource addresses and destination addresses of the network packages, attacks can be divided into four cases, which are respectively one host-one host, one host-many hosts, many hosts-one host and many hosts-many hosts. Specifically, the granular computing method is applied in intrusion detection. A cooperative intrusion detection model is proposed based on granular computing. The construction for an intrusion detection agent is presented.

SBD (Shot Boundary Detection) have great impact on effective browsing and retrieving of the video. It serves as the preliminary step to construct the content of videos. The Spatio-temoral local principle is used to detect shot boundary of videos. Based on careful analysis about the feature of videos, variable thresholds are used in different videos. A cooperative SBD model is proposed in the text. The components of the model are showed. The experimental results reveal the effectiveness and robustness of our method.

Because the network connection information contains nominal and linear attributes, and linear attributes are divided into continuous and discrete attributes, the network connection information is the heterogeneous data. The heterogeneous distance functions are used to cluster data in this paper. The cooperative network intrusion detection based on semi-supervised clustering algorithm is proposed. Firstly, the network data flows are divided into three data flows (TCP flow, UDP flow, and ICMP flow) according to network protocol and are sent to three detection agents. Then every detection agent constructs the detection model using the fuzzy c-means clustering algorithm based on the HVDM (Heterogeneous Value Difference Metric) distance. Finally, revise and verify the detection model by using test data. Simulation experiments are done by using KDD CUP 1999 data set, results show that the method presented here is feasible and efficient.

Based on insertion, Quick-Sort and Merge-Sort algorithms, this paper proposes an improved method about indexing and presents its corresponding parallel algorithm. Introduction of index table increases memory consumption but decreases consumption of record movement in sorting. The experiment demonstrates that executing CPU time of indexingbased sort algorithm is evidently less than that of other sort algorithms. Based on index table and parallel computing, the Merge-Sort algorithm saved the waiting and disposal time in which every two sub-merging sequences are sorted in single processor computer. This obtained better efficiency than the original Merge-Sort algorithm.

A first step required to allow video indexing and retrieval of visual data is to perform a temporal segmentation, that is, to find the location of camerashot transitions, which can be either abrupt or gradual. After a critical review of most approaches seeking to solve this problem, we propose a general method for video temporal segmentation. We adopt cooperative model to decide whether a shot transition exists or not within a given video sequence. A segmentation task can be completed through cooperative mechanism. Further, we apply fusion strategy to analyze candidate results and generate final results. The proposed method is evaluated on the TRECVID-2005 benchmarking platform and the experimental results reveal the effectiveness of the method.

Researchers have done lots of work in scan attack detection. Various methods have been proposed. Although these methods can defense some scan attacks from hackers in some degree, there are lots of missing detections and false alerts. Especially current intrusion detection systems are difficult to satisfy the demand of large-scale distributed network. After we carefully research on network topological architecture and scan attack method and mechanism, we find that scan attack always happened at network layer and transport layer. Then we propose a scan detection method based on distributed cooperative model. It is composed of feature-based detection, scenario-based detection and statistic-based detection. The experiment results show that this method has obvious advantages. It can efficiently detect more scan attacks.

This paper discusses that Role-Based Management (RBM) is applied in a Web-based teaching system. A user may have multiple roles. A role may be given to multiple users. Different roles have different privileges. From the implementation, it is found that roles are good mechanisms in designing a learning system.

Workflow management system can offer enterprises a method to implement better management. This paper focuses on existing workflow management system bugs in the collaboration, discusses the combinability between workflow management system and CSCW, puts forward a work-strategy in the workflow collaboration and a system reference model, and designs a CSCW-based workflow management system. This system is based on a knowledge library, uses the rules and knowledge in the knowledge library to realize the workflow collaboration. This system also applies in the area of stock so as to effectively solve the problem that in the current stock system the workflows cannot work in the collaboration.

When a new intrusion means is developed, many intrusion methods can be derived by exchanging the command sequences or by replacing commands with the functionally similar commands, which makes the detection of the developed intrusion very difficult. To overcome this problem, a cooperative intrusion detection model based on scenario is proposed, which is consisted of 5 layers. Topological order, isomorphic transformation and state transition analysis method are applied in the text. For an intrusion case we generate all the possible derived intrusions as an intrusion base. Based on this intrusion base, we present an efficient method to detect such intrusions by using finite automaton. Further, we apply data fusion to analysis suspicious data. A derived intrusion can be seen as an unknown intrusion, in this sense the technique presented in this paper can detect some unknown intrusions.

Making the workflow work in coordination is an important method to improve the efficiency of the workflow. This article has proposed a management method based on the knowledge library to make the workflows cooperative. This management method cooperatively controls the workflow from the tripartite aspects, which are the fixed flow, the special flow and the unknown flow. It advises to improve the cooperative degree of the fix flow through the knowledge which has already existed. It also suggests the coordination plan by way of inferring the special flow and the unknown flow so as to reduce manual intervention as far as possible. It can also carry on the revision and the intelligent increases to the knowledge library and realize the flexibility of the cooperative workflow scheme, so as to satisfy the enterprise's dynamic management request which brought by the business and organization's changes

The source data under the cooperated multi-agent work is always redundant, repetitive, inaccurate, uncompleted or inconsistent, so it is necessary to check, integrate and reduce the data, and convert the data into the evidences according with the internal reasoning mechanism. This paper presents the data pre-processing method and association analysis method for the alarming data.