With the widespread use of Graph Neural Networks (GNNs) for representation learning from network data, the fairness of GNN models has raised great attention lately. Fair GNNs aim to ensure that node representations can be accurately classified, but not easily associated with a specific group. Existing advanced approaches essentially enhance the generalisation of node representation in combination with data augmentation strategy and do not directly impose constraints on the fairness of GNNs. In this work, we identify that a fundamental reason for the unfairness of GNNs is the phenomenon of social homophily , i.e., users in the same group are more inclined to congregate. The message-passing mechanism of GNNs can cause users in the same group to have similar representations due to social homophily, leading model predictions to establish spurious correlations with sensitive attributes. Inspired by this reason, we propose a method called Equity-Aware GNN (EAGNN) towards fair graph representation learning. Specifically, to ensure that model predictions are independent of sensitive attributes while maintaining prediction performance, we introduce constraints for fair representation learning based on three principles: sufficiency, independence and separation. We theoretically demonstrate that our EAGNN method can effectively achieve group fairness. Extensive experiments on three datasets with varying levels of social homophily illustrate that our EAGNN method achieves the state-of-the-art performance across two fairness metrics and offers competitive effectiveness.

Hyper-relational knowledge graphs (HKGs) extend the traditional triplet-based knowledge graph by adding qualifiers to the relationships, making HKGs particularly useful for tasks that require more profound understanding and inference from relationships between entities. However, existing hyper-relational knowledge representation learning methods (HKRL) focus on direct neighbourhood information of entities only by neglecting the relational similarity of the main triple in hyper-relational facts and the attribute details in the qualifiers. In addition, few works extract common and private information across multiple views to minimize noise and interference. This paper proposes a multi-hypergraph disentanglement method for HKRL to address the above issues. Specifically, we first construct four hypergraphs to mine and utilise the inherent structure information of HKGs, and then propose to extract common representations among hypergraphs and private representations within individual hypergraphs to mine the semantic information and the task-relevant information, respectively. Experiment results on four real datasets demonstrate the effectiveness of the proposed method compared to SOTA methods in link prediction tasks on HKGs.

Anomaly detection in attributed graph data has been widely applied in real applications. However, the intricate topology of graph data, high-dimensional attributes, and class imbalance inherent in anomaly detection tasks render attributed graph anomaly detection a challenging task. To detect anomalies using the intricate topology information of graph data, a dual-masked autoencoders is proposed for attributed graph anomaly detection, denoted as MAGAD. Specifically, in the MAGAD, the class imbalance in attributed graph data is dealt with by randomly masking the original graph data to obtain masked graph data for the anomaly detection task. And then, a latent representation of the graph data is obtained by training dual autoencoders, where one autoencoder is developed for reconstructing the original graph data, and another for reconstructing randomly masked graph data. This assists in identifying abnormal nodes in the attributed graph data. Subsequently, to capture anomalous information from relevant features, MAGAD uses a random re-masking strategy for latent representations learned from the masked graph. Finally, the anomaly scores of the nodes are calculated using the learned latent representations from the decoders of the dual autoencoders. Experimental results on five real-world datasets demonstrate that the MAGAD algorithm outperforms state-of-the-art anomaly detection algorithms.

Increasing multiple behavior recommendation models have achieved great successes. However, many models do not consider commonalities and differences between behaviors and data sparsity of the target behavior. This article proposes a novel multi-behavior recommendation model based on contrastive clustering learning (MBRCC). Specifically, the graph convolutional network (GCN) is employed to obtain the embeddings of users and items, respectively. Then, three kinds of tasks (including behavior-level embedding, instance-level embedding, and cluster-level embedding) are designed to optimize the embeddings of users and items. In behavior-level embedding, we design an adaptive parameter learning strategy to analyze the impact of auxiliary behaviors on the target behavior. Then, the embeddings of users for each behavior are weighted to obtain the final embeddings of users. In instance-level embedding, we employ contrastive learning to analyze the instances of user and item for mitigating the issue of data sparsity. In cluster-level embedding, we design a new cluster contrastive learning method to capture the similarity between groups of user and item. Finally, we combine these three tasks to improve the quality of the embeddings of users and items. We conduct extensive experiments on three real-world datasets and experimental results indicate that the MBRCC remarkably outperforms numerous existing recommendation models.

The rapid development of Large Language Models (LLMs) creates new opportunities for recommender systems, especially by exploiting the side information (e.g., descriptions and analyses of items) generated by these models. However, aligning this side information with collaborative information from historical interactions poses significant challenges. The inherent biases within LLMs can skew recommendations, resulting in distorted and potentially unfair user experiences. On the other hand, propensity bias causes side information to be aligned in such a way that it often tends to represent all inputs in a low-dimensional subspace, leading to a phenomenon known as dimensional collapse, which severely restricts the recommender system’s ability to capture user preferences and behaviors. To address these issues, we introduce a novel framework named Counterfactual LLM Recommendation (CLLMR). Specifically, we propose a spectrum-based side information encoder that implicitly embeds structural information from historical interactions into the side information representation, thereby circumventing the risk of dimension collapse. Furthermore, our CLLMR approach explores the causal relationships inherent in LLM-based recommender systems. By leveraging counterfactual inference, we counteract the biases introduced by LLMs. Extensive experiments demonstrate that our CLLMR approach consistently enhances the performance of various recommender models.

Recommender systems rely on observational data to predict user ratings for unseen items. Since the observational data is typically missing not at random (MNAR), they contain biases, predominantly selection bias, thus models trained on such data are inherently biased. If a shadow variable, which is a variable used instead of user's latent variables that influence both the treatment and the outcome, can be identified, it is possible to build unbiased models for recommender systems. To overcome the challenge of manually identifying valid shadow variables in the data, we propose a Shadow Variable Catcher (SVC), a model designed to learn the latent representation of shadow variables from observational data. By using the learned representation of shadow variables by SVC, we develop the Shadow Debiased Recommender (SDR) method to build an unbiased collaborative filtering model for addressing selection bias in recommender systems. Comprehensive experiments on both synthetic and real-world datasets, have verified the performance of SDR and demonstrated its effectiveness and robustness, and offer new insights into the mitigation of bias in recommender systems.

The spread of misinformation on social media is a serious issue that can have negative consequences for public health and political stability. While detecting and identifying misinformation can be challenging, many attempts have been made to address this problem. However, traditional models that focus on pairwise relationships on misinformation propagation paths may not be effective in capturing the underlying connections among multiple tweets. To address this limitation, the proposed “Conversation-Branch-Tweet” hypergraph convolutional network (CBT-HGCN) uses a hypergraph to represent the internal structure and content of tweet data, with tweets and their replies viewed as nodes and hyperedges, respectively. The model first pre-processes the tweets of a conversation and then uses a pre-trained model as an encoder to extract node information. Finally, a hypergraph convolution network is used as an information fuser for classification. Experimental results on three benchmark datasets (Twitter15, Twitter16, and Pheme) show that the proposed model outperforms several strong baseline models and achieves state-of-the-art performance. This indicates that the CBT-HGCN approach is effective in detecting and identifying misinformation on social media by capturing the underlying connections among multiple tweets.

Recommendation has become especially crucial during the COVID-19 pandemic as a significant number of people rely on online shopping from home. Existing recommendation algorithms, designed to address issues like cold start and data sparsity, often overlook the time constraints of users. Specifically, users expect to receive recommendations for products of interest in the shortest possible time. To address this challenge, we propose a novel collaborative filtering recommendation algorithm that leverages the advantages of quantum computing circuits based on data reconstruction. This approach allows for the rapid identification of users similar to the target user, thereby improving recommendation speed. In our method, we utilize the information of known users to linearly reconstruct that of the target users, forming a relational matrix. Subsequently, we employ \(l_{2,1}-\) norm and \(l_{1}-\) norm to sparsely constrain the relationship matrix, deducing the weight of each known user. The final step involves providing similar recommendations to target users based on these weights. Furthermore, we implement the proposed algorithm using a quantum circuit, enabling exponential acceleration. The final weight matrix is derived from the quantum state outputted by the circuit. The speed of this process is theoretically demonstrated in detail. Experimental results indicate that our algorithm outperforms state-of-the-art methods in terms of root mean squared error (RMSE), mean absolute error (MAE) and normalized discounted cumulative gain (NDCG). Compared to state-of-the-art comparison algorithms, the proposed algorithm achieves the fastest recommendation speed across eight public datasets.

It has become a cardinal task to identify fake information (misinformation) on social media, because it has significantly harmed the government and the public. There are many spam bots maliciously retweeting misinformation. This study proposes an efficient model for detecting misinformation with self-supervised contrastive learning. A B ayesian graph L ocal extrema C onvolution (BLC) is first proposed to aggregate node features in the graph structure. The BLC approach considers unreliable relationships and uncertainties in the propagation structure, and the differences between nodes and neighboring nodes are emphasized in the attributes. Then, a new long-tail strategy for matching long-tail users with the global social network is advocated to avoid over-concentration on high-degree nodes in graph neural networks. Finally, the proposed model is experimentally evaluated with two public Twitter datasets and demonstrates that the proposed long-tail strategy significantly improves the effectiveness of existing graph-based methods in terms of detecting misinformation. The robustness of BLC has also been examined on three graph datasets and demonstrates that it consistently outperforms traditional algorithms when perturbed by 15% of a dataset.

Incomplete multi-view clustering has represented a significant role in grouping real images. In this study, a novel robust tensor subspace learning (RTSL) is proposed for incomplete multi-view clustering. Specifically, the missing samples within views are first recovered by matrix factorization. The recovered information is utilized for latent representations learning. And then, the obtained latent representations are organized from all views into a third-order tensor and the intrinsic sample relations are captured with tensor linear representation. Moreover, a low-rank sample coefficient tensor is sought to capture high-order connections among views by imposing the tensor nuclear norm. Compared with traditional learning paradigms in the vector space, the sample relations within each view as well as across views could be preserved with the aid of robust tensor subspace learning. As a result, our model can simultaneously handle the missing samples and exploit the intrinsic correlations, leading to enhanced representation capability and better quality of the recovered data. We design an efficient iterative optimization strategy to solve the proposed method. Experimental results on eight datasets show that our model outperforms other competing approaches.

Learning to hash is of particular interest in information retrieval for large-scale data due to its high efficiency and effectiveness. Most studies in hashing concentrate on constructing new hashing models, but rarely touch the correlation and redundancy between hash bits derived. In this article, we first introduce a general schema of hash bit reduction to derive compact and informative binary codes for hashing techniques. Further, we take locality sensitive hashing, one of the most widely-used hashing methods, as an example and propose a novel and two-stage binary code refinement method under the reduction schema. Specifically, the proposed method includes two stages, i.e., bit evaluation and bit refinement. The former stage aims to initially extract a small portion of informative hash bits in terms of their importance and quality evaluated by bit balance and similarity preservation. Then, the representation capabilities of the reduced hash bits are strengthened further by refining their binary values. The purpose of refinement is to lessen the correlations and redundancies between the reduced bits, making themselves more discriminative. The experimental results on three widely-used data collections confirm the effectiveness of the proposed bit reduction method and its superiority over the state-of-the-art hashing methods, as well as a bit selection method.

The challenge of the dimensional disaster in multi-view data is an ongoing and formidable issue. Current multi-view feature selection algorithms aim to reduce dimensions by learning a feature subset that effectively captures the overall information of the data, integrating the characteristics from multiple views. However, they often overlook the detrimental impact of noise in the data, which compromises the performance of multi-view feature selection and yields inefficient feature subsets. To address this problem, this paper proposes an anti-noise multi-view feature selection algorithm. In particular, we begin by combining least squares loss and regularization techniques to learn the relationship between the data and labels. Subsequently, we introduce sample constraints, including view weight and sample weight, as well as feature weight factors, into the objective function. This incorporation reduces the significance of noisy samples, thereby enhancing the algorithm’s ability to resist noise interference. In comparative evaluations with state-of-the-art algorithms, the proposed algorithm exhibits an average improvement of 4.42% in classification accuracy when applied to publicly available datasets with added noise1

It is challenging to detect 3D objects from noise point clouds by Graph Neural Networks (GNNs), though graph-based methods have shown promising results in 3D classifications. Since strong robustness against noise is offered by hypergraph, a relative paradigm named HyperGraph Construction-Compression-Conversion (HG3C) is proposed for detecting 3D objects from noise point clouds. Our method presents the capacity of reducing graph redundancy and capturing the variances from multiple features, by pre-encoding the graph, to improve the graph representations in point clouds. A fused graph neural network is further designed to predict the shape and category of the target in converted graphs. The experiments, on both the KITTI and Nuscene, show that the proposed approach achieves leading accuracy. Our results demonstrate the potential of using the hypergraph transformation to extract and compress point cloud information from noisy point clouds.

Image copy detection is an important technology of copyright protection. This paper proposes an efficient hashing method for image copy detection using 2D-2D (two-directional two-dimensional) PCA (Principal Component Analysis). The key is the discovery of the translation invariance of 2D-2D PCA. With the property of translation invariance, a novel model of extracting rotation-invariant low-dimensional features is designed by combining PCT (Polar Coordinate Transformation) and 2D-2D PCA. The PCT can convert an input rotated image to a translation matrix. Since the 2D-2D PCA is invariant to translation, the low-dimensional features learned from the translation matrix are rotation-invariant. Moreover, vector distances of low-dimensional features are stable to common digital operations and thus hash construction with the vector distances is of robustness and compactness. Three open image datasets are exploited to conduct various experiments for validating efficiencies of the proposed method. The results demonstrate that the proposed method is much better than some representative hashing methods in the performances of classification and copy detection.

In linguistics, the uncertainty of context due to polysemy is widespread, which attracts much attention. Quantum-inspired complex word embedding based on Hilbert space plays an important role in natural language processing (NLP), which fully leverages the similarity between quantum states and word tokens. A word containing multiple meanings could correspond to a single quantum particle which may exist in several possible states, and a sentence could be analogous to the quantum system where particles interfere with each other. Motivated by quantum-inspired complex word embedding, interpretable complex-valued word embedding (ICWE) is proposed to design two end-to-end quantum-inspired deep neural networks (ICWE-QNN and CICWE-QNN representing convolutional complex-valued neural network based on ICWE) for binary text classification. They have the proven feasibility and effectiveness in the application of NLP and can solve the problem of text information loss in CE-Mix [1] model caused by neglecting the important linguistic features of text, since linguistic feature extraction is presented in our model with deep learning algorithms, in which gated recurrent unit (GRU) extracts the sequence information of sentences, attention mechanism makes the model focus on important words in sentences and convolutional layer captures the local features of projected matrix. The model ICWE-QNN can avoid random combination of word tokens and CICWE-QNN fully considers textual features of the projected matrix. Experiments conducted on five benchmarking classification datasets demonstrate our proposed models have higher accuracy than the compared traditional models including CaptionRep BOW, DictRep BOW and Paragram-Phrase, and they also have great performance on F1-score. Eespecially, CICWE-QNN model has higher accuracy than the quantum-inspired model CE-Mix as well for four datasets including SST, SUBJ, CR and MPQA. It is a meaningful and effictive exploration to design quantum-inspired deep neural networks to promote the performance of text classification.

A quantum circuit learning approach is studied to carry out the fast-fitting of Gaussian functions. First, a parameterized structure is designed for quantum circuits based on the boson sampling model. And then, the training procedure of exploiting gradient-based optimizations is presented to iteratively update the gradient of the loss function concerning circuit parameters. For efficiency, two kinds of circuit loss, the kernel maximum mean discrepancy and the mean absolute error, are used in the training procedure, which are both competent to achieve quantum circuit learning well. It is significant that the two circuit losses assist in reducing the variance to$2.54 \times 10^{-6}$and$6.91 \times 10^{-6}$, respectively. Finally, a kind of quantum circuit fixed structure is developed with the boson sampling model that can decrease the model complexity as the circuit depth d grows. Sets of experiments have been conducted to evaluate the proposed quantum circuit learning scheme, and demonstrate that our parameterized approach is efficient and promising, and it is worth looking forward to solving practical application problems with quantum computers since valid quantum circuits for Gaussian function fast-fitting can be designed indeed.

Data collected from nature is usually unlabeled, and it is difficult to be used directly. This issue is well addressed by crowdsourcing, which provides a reasonable way for effectively using these unlabeled data. Generally, workers in crowdsourcing tasks are not professionals, so it is hard to obtain high-quality labels. To address this issue, a robust sparse weighted classification algorithm is proposed, which try to adjust the samples that are not correctly classified in the original lables as much as possible. Specifically, we evalute the ability of different workers(indicator weight matrix) to accurately label different samples by fitting the real data matrix to its weighted reconstruction matrix. And then,$ l_{2,1}$-norm and worker labeling ability similarity matrix are added, and negative effects of some bad workers are eliminated through the row sparsity property of$ l_{2,1}$-norm. Finally, the optimal indicator weight matrix is obtained by optimizing the two matrices in the objective function simultaneously. Therefore, the obtained optimal indicator weight matrix takes the similarity of worker labeling ability into consideration, and infers all the predicted labels. The results on synthetic and real data sets demonstrate that our algorithm is superior to other state-of-the-art methods.

KNN classification is an improvisational learning mode, in which they are carried out only when a test data is predicted that set a suitable K value and search the K nearest neighbors from the whole training sample space, referred them to the lazy part of KNN classification. This lazy part has been the bottleneck problem of applying KNN classification due to the complete search of K nearest neighbors. In this paper, a one-step computation is proposed to replace the lazy part of KNN classification. The one-step computation actually transforms the lazy part to a matrix computation as follows. Given a test data, training samples are first applied to fit the test data with the least squares loss function. And then, a relationship matrix is generated by weighting all training samples according to their influence on the test data. Finally, a group lasso is employed to perform sparse learning of the relationship matrix. In this way, setting K value and searching K nearest neighbors are both integrated to a unified computation. In addition, a new classification rule is proposed for improving the performance of one-step KNN classification. The proposed approach is experimentally evaluated, and demonstrated that the one-step KNN classification is efficient and promising.

Distance function is a main metrics of measuring the affinity between two data points in machine learning. Extant distance functions often provide unreachable distance values in real applications. This can lead to incorrect measure of the affinity between data points. This paper proposes a reachable distance function for KNN classification. The reachable distance function is not a geometric direct-line distance between two data points. It gives a consideration to the class attribute of a training dataset when measuring the affinity between data points. Concretely speaking, the reachable distance between data points includes their class center distance and real distance. Its shape looks like “Z,” and we also call it a Z distance function. In this way, the affinity between data points in the same class is always stronger than that in different classes. Or, the intraclass data points are always closer than those interclass data points. We evaluated the reachable distance with experiments, and demonstrated that the proposed distance function achieved better performance in KNN classification.

Cross-modal hashing is a hot issue in the multimedia community, which is to generate compact hash code from multimedia content for efficient cross-modal search. Two challenges, i.e., (1) How to efficiently enhance cross-modal semantic mining is essential for cross-modal hash code learning, and (2) How to combine multiple semantic correlations learning to improve the semantic similarity preserving, cannot be ignored. To this end, this paper proposed a novel end-to-end cross-modal hashing approach, named Multiple Semantic Structure-Preserving Quantization (MSSPQ) that is to integrate deep hashing model with multiple semantic correlation learning to boost hash learning performance. The multiple semantic correlation learning consists of inter-modal and intra-modal pairwise correlation learning and Cosine correlation learning, which can comprehensively capture cross-modal consistent semantics and realize semantic similarity preserving. Extensive experiments are conducted on three multimedia datasets, which confirms that the proposed method outperforms the baselines.

The stock market investors aim at maximizing their investment returns. Stock recommendation task is to recommend stocks with higher return ratios for the investors. Most stock prediction methods study the historical sequence patterns to predict stock trend or price in the near future. In fact, the future price of a stock is correlated not only with its historical price, but also with other stocks. In this article, we take into account the relationships between stocks (corporations) by stock relation graph. Furthermore, we propose a Time-aware Relational Attention Network (TRAN) for graph-based stock recommendation according to return ratio ranking. In TRAN, the time-aware relational attention mechanism is designed to capture time-varying correlation strengths between stocks by the interaction of historical sequences and stock description documents. With the dynamic strengths, the nodes of the stock relation graph aggregate the features of neighbor stock nodes by graph convolution operation. For a given group of stocks, the proposed TRAN model can output the ranking results of stocks according to their return ratios. The experimental results on several real-world datasets demonstrate the effectiveness of our TRAN for stock recommendation.

The KNN algorithm is one of the most popular data mining algorithms. It has been widely and successfully applied to data analysis applications across a variety of research topics in computer science. This paper illustrates that, despite its success, there remain many challenges in KNN classification, including K computation, nearest neighbor selection, nearest neighbor search and classification rules. Having established these issues, recent approaches to their resolution are examined in more detail, thereby providing a potential roadmap for ongoing KNN-related research, as well as some new classification rules regarding how to tackle the issue of training sample imbalance. To evaluate the proposed approaches, some experiments were conducted with 15 UCI benchmark datasets.

Stego-images are often contaminated by interchannel noise or active noise attack when communicating on the Web. And it is challenging to restore embedded image from corrupted stego-image. This paper studies akNN-bit approximation algorithm to remove noises in embedded image. The proposed algorithm distinguishes reliable bits from extracted bits, and estimates pixel values by keeping reliable bits unchanged and correcting unreliable bits. Specifically, the 8th (highest) unreliable bit of a pixel can be approximated with its nearest neighbor pixels. And then, if an unreliable bit locates at any one of the$5^{th}\sim 7^{th}$bits of a pixel, it is adjusted with two nearest neighbors of the pixel, where the pixel is in-between these two nearest neighbors. Finally, for other unreliable bits, each one is approximated by the maximum and minimum possible values of nearest neighbors of its pixel. We conduct experiments for illustrating the efficiency, and demonstrate that the proposed algorithm can recover the embedded images with good visual quality from corrupted stego-images.

Unsupervised spectral feature selection (USFS) methods could output interpretable and discriminative results by embedding a Laplacian regularizer in the framework of sparse feature selection to keep the local similarity of the training samples. To do this, USFS methods usually construct the Laplacian matrix using either a general-graph or a hyper-graph on the original data. Usually, a general-graph could measure the relationship between two samples while a hyper-graph could measure the relationship among no less than two samples. Obviously, the general-graph is a special case of the hyper-graph and the hyper-graph may capture more complex structure of samples than the general graph. However, in previous USFS methods, the construction of the Laplacian matrix is separated from the process of feature selection. Moreover, the original data usually contain noise. Each of them makes difficult to output reliable feature selection models. In this paper, we propose a novel feature selection method by dynamically constructing a hyper-graph based Laplacian matrix in the framework of sparse feature selection. Experimental results on real datasets showed that our proposed method outperformed the state-of-the-art methods in terms of both clustering and segmentation tasks.

Cross-modal hashing is a critical but very challenging task that is to retrieve similar samples of one modality via queries of other modalities. To improve the unsupervised cross-modal hashing, domain adaptation techniques can be used to support unsupervised hashing learning by transferring semantic knowledge from labeled source domain to unlabeled target domain. However, there are two problems that cannot be ignored: (1) most of domain adaptation based researches mainly focused on unimodal hashing or cross-modal real value-based retrieval but the study for cross-modal hashing is limited; (2) most existing studies only consider one or two consistency constraints during the domain adaptation learning. To this end, this paper propose a novel end-to-end framework to realize unsupervised domain adaptation for cross-modal hashing. This method, dubbed M$^2$GUDA, including four different consistency constraints: structure consistency, domain consistency, semantic consistency and modality consistency for domain adaptation learning. Besides, to enhance the structure consistency learning, we develop a multi-metrics graph modeling method to capture structure information comprehensively. Extensive experiments are performed on three common used benchmarks to evaluate the effectivity of our method. The results show that our method outperforms several state-of-the-art cross-modal hashing methods.

Similarity representation plays a central role in increasingly popular anomaly detection techniques, which have been successfully applied in various realistic scenes. Until now, many low-rank representation techniques have been introduced to measure the similarity relations of data; yet, they only concern to minimize reconstruction errors, without involving the structural information of data. Besides, the traditional low-rank representation methods often take nuclear norm as their low-rank constraints, easily yielding a suboptimal solution. To address the problems above, in this article, we propose a novel anomaly detection method, which exploits kernel preserving embedding, as well as the double nuclear norm, to explore the similarity relations of data. Based on the similarity relations, a kind of probability transition matrix is derived, and a tailored random walk is further adopted to reveal anomalies. The proposed method can not only preserve the manifold structural properties of the data, but also alleviate the suboptimal problem. To validate the superiority of our method, extensive experiments with eight popular anomaly detection algorithms were conducted on 12 widely used datasets. The experimental results show that our detection method outperformed the state-of-the-art anomaly detection algorithms in most cases.

Different from previous imputation methods which impute missing values in the incomplete samples by using the information in the complete samples, this paper proposes a Date-drive Incremental imputation Model, DIM for short, which uses all available information in the data set to impute missing values economically, effectively, orderly, and iteratively. To this end, we propose a scoring rule to rank the missing features by taking into account both the economical criterion and the effective imputation information. The economical criterion takes both the imputation cost and the discriminative ability of the feature into account, while the effective imputation information enables to use all observed information in the data set including the imputed missing values to impute the left missing values. During the imputation process, our DIM first detects the neednot-impute samples for reducing the imputation cost and noise, and then selects the missing features with the top rank to impute first. The imputation process orderly imputes the missing features until all missing values are imputed or the imputation cost is exhausted. Experimental results on UCI data sets demonstrated the advantages of our proposed DIM, compared to the comparison methods, in terms of prediction accuracy and classification accuracy.

It has been demonstrated that the segmentation performance is highly dependent on both subspace preservation and graph connectivity. In the literature, the full connectivity method linearly represents each data point ( e.g., a pixel in one image) by all data points for achieving subspace preservation, while the sparse connectivity method was designed to linearly represent each data point by a set of data points for achieving graph connectivity. However, previous methods only focused on either subspace preservation or graph connectivity. In this article, we propose a Sparse Graph Connectivity (SGC) method for image segmentation to automatically learn the affinity matrix from the low-dimensional space of original data, which aims at simultaneously achieving subspace preservation and graph connectivity. To do this, the proposed SGC simultaneously learns a self-representation affinity matrix for subspace preservation and a sparse affinity matrix for graph connectivity, from the intrinsic low-dimensional feature space of high-dimensional original data. Meanwhile, the self-representation affinity matrix is pushed to be similar to the sparse affinity as well as be the final segmentation results. Experimental result on synthetic and real-image datasets showed that our SGC method achieved the best segmentation performance, compared to state-of-the-art segmentation methods.

Since the data in each view may contain distinct information different from other views as well as has common information for all views in multi-view learning, many multi-view clustering methods have been designed to use these information (including the distinct information for each view and the common information for all views) to improve the clustering performance. However, previous multi-view clustering methods cannot effectively detect these information so that difficultly outputting reliable clustering models. In this article, we propose a fuzzy, sparse, and robust multi-view clustering method to consider all kinds of relations among the data (such as view importance, view stability, and view diversity), which can effectively extract both distinct information and common information as well as balance these two kinds of information. Moreover, we devise an alternating optimization algorithm to solve the resulting objective function as well as prove that our proposed algorithm achieves fast convergence. It is noteworthy that existing multi-view clustering methods only consider a part of the relations, and thus are a special case of our proposed framework. Experimental results on synthetic datasets and real datasets show that our proposed method outperforms the state-of-the-art clustering methods in terms of evaluation metrics of clustering such as clustering accuracy, normalized mutual information, purity, and adjusted rand index.

This paper presents a new image hashing that is designed with tensor decomposition (TD), referred to as TD hashing, where image hash generation is viewed as deriving a compact representation from a tensor. Specifically, a stable three-order tensor is first constructed from the normalized image, so as to enhance the robustness of our TD hashing. A popular TD algorithm, called Tucker decomposition, is then exploited to decompose the three-order tensor into a core tensor and three orthogonal factor matrices. As the factor matrices can reflect intrinsic structure of original tensor, hash construction with the factor matrices makes a desirable discrimination of the TD hashing. To examine these claims, there are 14,551 images selected for our experiments. A receiver operating characteristics (ROC) graph is used to conduct theoretical analysis and the ROC comparisons illustrate that the TD hashing outperforms some state-of-the-art algorithms in classification performance between the robustness and discrimination.

Previous multi-view spectral clustering methods are a two-step strategy, which first learns a fixed common representation (or common affinity matrix) of all the views from original data and then conducts k-means clustering on the resulting common affinity matrix. The two-step strategy is not able to output reasonable clustering performance since the goal of the first step (i.e., the common affinity matrix learning) is not designed for achieving the optimal clustering result. Moreover, the two-step strategy learns the common affinity matrix from original data, which often contain noise and redundancy to influence the quality of the common affinity matrix. To address these issues, in this paper, we design a novel One-step Multi-view Spectral Clustering (OMSC) method to output the common affinity matrix as the final clustering result. In the proposed method, the goal of the common affinity matrix learning is designed to achieving optimal clustering result and the common affinity matrix is learned from low-dimensional data where the noise and redundancy of original high-dimensional data have been removed. We further propose an iterative optimization method to fast solve the proposed objective function. Experimental results on both synthetic datasets and public datasets validated the effectiveness of our proposed method, comparing to the state-of-the-art methods for multi-view clustering.

Traditional graph clustering methods consist of two sequential steps, i.e., constructing an affinity matrix from the original data and then performing spectral clustering on the resulting affinity matrix. This two-step strategy achieves optimal solution for each step separately, but cannot guarantee that it will obtain the globally optimal clustering results. Moreover, the affinity matrix directly learned from the original data will seriously affect the clustering performance, since high-dimensional data are usually noisy and may contain redundancy. To address the above issues, this paper proposes a Low-rank Sparse Subspace (LSS) clustering method via dynamically learning the affinity matrix from low-dimensional space of the original data. Specifically, we learn a transformation matrix to project the original data to their low-dimensional space, by conducting feature selection and subspace learning in the sample self-representation framework. Then, we utilize the rank constraint and the affinity matrix directly obtained from the original data to construct a dynamic and intrinsic affinity matrix. Moreover, each of these three matrices is updated iteratively while fixing the other two. In this way, the affinity matrix learned from the low-dimensional space is the final clustering results. Extensive experiments are conducted on both synthetic and real datasets to show that our proposed LSS method outperforms the state-of-the-art clustering methods.

This paper proposes a new unsupervised spectral feature selection method to preserve both the local and global structure of the features as well as the samples. Specifically, our method uses the self-expressiveness of the features to represent each feature by other features for preserving the local structure of features, and a low-rank constraint on the weight matrix to preserve the global structure among samples as well as features. Our method also proposes to learn the graph matrix measuring the similarity of samples for preserving the local structure among samples. Furthermore, we propose a new optimization algorithm to the resulting objective function, which iteratively updates the graph matrix and the intrinsic space so that collaboratively improving each of them. Experimental analysis on 12 benchmark datasets showed that the proposed method outperformed the state-of-the-art feature selection methods in terms of classification performance.

The K Nearest Neighbor (kNN) method has widely been used in the applications of data mining and machine learning due to its simple implementation and distinguished performance. However, setting all test data with the same k value in the previous kNN methods has been proven to make these methods impractical in real applications. This article proposes to learn a correlation matrix to reconstruct test data points by training data to assign different k values to different test data points, referred to as the Correlation Matrix kNN (CM-kNN for short) classification. Specifically, the least-squares loss function is employed to minimize the reconstruction error to reconstruct each test data point by all training data points. Then, a graph Laplacian regularizer is advocated to preserve the local structure of the data in the reconstruction process. Moreover, an ℓ 1 -norm regularizer and an ℓ 2, 1 -norm regularizer are applied to learn different k values for different test data and to result in low sparsity to remove the redundant/noisy feature from the reconstruction process, respectively. Besides for classification tasks, the kNN methods (including our proposed CM-kNN method) are further utilized to regression and missing data imputation. We conducted sets of experiments for illustrating the efficiency, and experimental results showed that the proposed method was more accurate and efficient than existing kNN methods in data-mining applications, such as classification, regression, and missing data imputation.

This paper designs an efficient image hashing with a ring partition and a nonnegative matrix factorization (NMF), which has both the rotation robustness and good discriminative capability. The key contribution is a novel construction of rotation-invariant secondary image, which is used for the first time in image hashing and helps to make image hash resistant to rotation. In addition, NMF coefficients are approximately linearly changed by content-preserving manipulations, so as to measure hash similarity with correlation coefficient. We conduct experiments for illustrating the efficiency with 346 images. Our experiments show that the proposed hashing is robust against content-preserving operations, such as image rotation, JPEG compression, watermark embedding, Gaussian low-pass filtering, gamma correction, brightness adjustment, contrast adjustment, and image scaling. Receiver operating characteristics (ROC) curve comparisons are also conducted with the state-of-the-art algorithms, and demonstrate that the proposed hashing is much better than all these algorithms in classification performances with respect to robustness and discrimination.

This paper takes manifold learning and regression simultaneously into account to perform unsupervised spectral feature selection. We first extract the bases of the data, and then represent the data sparsely using the extracted bases by proposing a novel joint graph sparse coding model, JGSC for short. We design a new algorithm TOSC to compute the resulting objective function of JGSC, and then theoretically prove that the proposed objective function converges to its global optimum via the proposed TOSC algorithm. We repeat the extraction and the TOSC calculation until the value of the objective function of JGSC satisfies pre-defined conditions. Eventually the derived new representation of the data may only have a few non-zero rows, and we delete the zero rows (a.k.a. zero-valued features) to conduct feature selection on the new representation of the data. Our empirical studies demonstrate that the proposed method outperforms several state-of-the-art algorithms on real datasets in term of the kNN classification performance.

One of the challenges in data mining is the dimensionality of data, which is often very high and prevalent in many domains, such as text categorization and bio-informatics. The high-dimensionality of data may bring many adverse situations to traditional learning algorithms. To cope with this issue, feature selection has been put forward. Currently, many efforts have been attempted in this field and lots of feature selection algorithms have been developed. In this paper we propose a new selection method to pick discriminative features by using information measurement. The main characteristic of our selection method is that the selection procedure works like feature clustering in a hierarchically agglomerative way, where each feature is considered as a cluster and the between-cluster and within-cluster distances are measured by mutual information and the coefficient of relevancy respectively. Consequently, the final aggregated cluster is the selection result, which has the minimal redundancy among its members and the maximal relevancy with the class labels. The simulation experiments on seven datasets show that the proposed method outperforms other popular feature selection algorithms in classification performance.

Missing data imputation is a key issue in learning from incomplete data. Various techniques have been developed with great successes on dealing with missing values in data sets with homogeneous attributes (their independent attributes are all either continuous or discrete). This paper studies a new setting of missing data imputation, i.e., imputing missing data in data sets with heterogeneous attributes (their independent attributes are of different types), referred to as imputing mixed-attribute data sets. Although many real applications are in this setting, there is no estimator designed for imputing mixed-attribute data sets. This paper first proposes two consistent estimators for discrete and continuous missing target values, respectively. And then, a mixture-kernel-based iterative estimator is advocated to impute mixed-attribute data sets. The proposed method is evaluated with extensive experiments compared with some typical algorithms, and the result demonstrates that the proposed approach is better than these existing imputation methods in terms of classification accuracy and root mean square error (RMSE) at different missing ratios.

An association rule ArarrB is useful to predict that B will likely occur when A occurs. This is a classical association rule. In real world applications, such as bioinformatics and medical research, there are many follow correlations between itemsets A and B: B likely occurs n times after A occurred m times, wrote tom, BN>. We refer to this follow-correlation as P3.1 itemset-pairs because3, B1> like that in the example ( Example 2) should be uninterested in association analysis. This paper designs an efficient algorithm for identifying P3.1 itemset-pairs in sequential data. We experimentally evaluate our approach, and demonstrate that the proposed approach is efficient and promising.

A bridging rule in this paper has its antecedent and action from different conceptual clusters. We first design two algorithms for mining bridging rules between clusters in a database, and then propose two non-linear metrics for measuring the interestingness of bridging rules. Bridging rules can be distinct from association rules (or frequent itemsets). This is because (1) bridging rules can be generated by infrequent itemsets that are pruned in association rule mining; and (2) bridging rules are measured by the importance that includes the distance between two conceptual clusters, whereas frequent itemsets are measured by only the support.

Creating mappings between database schema and Web ontology is a preconditioning process in the generation of ontological annotations for dynamic Web page contents extracted from the database. In this paper, a practical approach to creating mappings between a relational database schema and an OWL ontology is presented. The approach can automatically construct the mappings by following a set of predefined heuristic rules based on the conceptual correspondences between the schema and the ontology. This automatic mapping is implemented as the core functionality in a prototype tool D2OMapper that has some assistant functions to help the user manually create and maintain the mappings. Case studies show that the proposed approach is effective and the produced mappings can be applied to semantic annotation of database-based, dynamic Web pages

Recent-biased approximations have received increased attention recently as a mechanism for learning trend patterns from time series or data streams. They have shown promise for clustering time series and incrementally pattern maintaining. In this paper, we design a generalized dimension-reduction framework for recent-biased approximations, aiming at making traditional dimension-reduction techniques actionable in recent-biased time series analysis. The framework is designed in two ways: equi-segmented scheme and vari-segmented scheme. In both schemes, time series data are first partitioned into segments and a dimension-reduction technique is applied to each segment. Then, more coefficients are kept for more recent data while fewer kept for older data. Thus, more details are preserved for recent data and fewer coefficients are kept for the whole time series, which improves the efficiency greatly. We experimentally evaluate the proposed approach, and demonstrate that traditional dimension-reduction techniques, such as SVD, DFT, DWT, PIP, PAA, and landmarks, can be embedded into our framework for recent-biased approximations over streaming time series.

Many real-world data sets for machine learning and data mining contain missing values and much previous research regards it as a problem and attempts to impute missing values before training and testing. In this paper, we study this issue in cost-sensitive learning that considers both test costs and misclassification costs. If some attributes (tests) are too expensive in obtaining their values, it would be more cost-effective to miss out their values, similar to skipping expensive and risky tests (missing values) in patient diagnosis (classification). That is, "missing is useful" as missing values actually reduces the total cost of tests and misclassifications and, therefore, it is not meaningful to impute their values. We discuss and compare several strategies that utilize only known values and that "missing is useful" for cost reduction in cost-sensitive decision tree learning.

This paper presents an efficient method for mining both positive and negative association rules in databases. The method extends traditional associations to include association rules of forms A ⇒ ¬ B , ¬ A ⇒ B , and ¬ A ⇒ ¬ B , which indicate negative associations between itemsets. With a pruning strategy and an interestingness measure, our method scales to large databases. The method has been evaluated using both synthetic and real-world databases, and our experimental results demonstrate its effectiveness and efficiency.

Many large organizations have multiple data sources, such as different branches of an interstate company. While putting all data together from different sources might amass a huge database for centralized processing, mining association rules at different data sources and forwarding the rules (rather than the original raw data) to the centralized company headquarter provides a feasible way to deal with multiple data source problems. In the meanwhile, the association rules at each data source may be required for that data source in the first instance, so association analysis at each data source is also important and useful. However, the forwarded rules from different data sources may be too many for the centralized company headquarter to use. This paper presents a weighting model for synthesizing high-frequency association rules from different data sources. There are two reasons to focus on high-frequency rules. First, a centralized company headquarter is interested in high-frequency rules because they are supported by most of its branches for corporate profitability. Second, high-frequency rules have larger chances to become valid rules in the union of all data sources. In order to extract high-frequency rules efficiently, a procedure of rule selection is also constructed to enhance the weighting model by coping with low-frequency rules. Experimental results show that our proposed weighting model is efficient and effective.

Current data mining techniques may not be helpful for mining some companies/organizations such as nuclear power plants and earthquake bureaus, which have only small databases. Apparently, these companies/organizations also expect to apply data mining techniques to extract useful patterns in their databases so as to make their decisions. However, data in these databases such as the accident database of a nuclear power plant and the earthquake database in an earthquake bureau, may not be large enough to form any patterns. To meet the applications, we present a new mining model in this paper, which is based on the collecting knowledge from such as Web, journals, and newspapers.