Search

Community detection is one of the fundamental tasks in graph mining, which aims to identify groups of nodes in complex networks. Recently, network embedding techniques have demonstrated their strong power in advancing the community detection task and achieve better performance than various traditional methods. Despite their empirical success, most of the existing algorithms directly leverage the observed coarse network structure for community detection. Therefore, they often lead to suboptimal performance as the observed connections fail to capture the essential tie strength information among nodes precisely and account for the impact of noisy links. In this paper, an optimal network structure for community detection is introduced to characterize the fine-grained tie strength information between connected nodes and alleviate the adverse effects of noisy links. To obtain an expressive node representation for community detection, we learn the optimal network structure and network embeddings in a joint framework, instead of using a two-stage approach to derive the node embeddings from the coarse network topology. In particular, we formulate the joint framework as an optimization problem and an alternating optimization algorithm is exploited to solve the proposed problem. Extensive experiments on both synthetic and real-world networks demonstrate the effectiveness and superiority of our framework.

Learning from imbalanced data streams differs from the traditional learning paradigm due to the issues of imbalanced classes. It has significant implications in a myriad of real-world applications, ranging from financial risk, network security, to medical diagnosis. Moreover, outliers usually appear in data streams. The issue of class imbalance or anomaly itself could negatively affect the performance of the underlying learning algorithms, and their combination makes the learning problem harder. In this work, we propose an anomaly detection aided budget online weighted learning method (BOW-LM) to identify positive and negative instances from imbalanced data streams. BOW-LM is based on the widely used Feedforward Networks with Random Weights. An agile lightweight anomaly detector is designed based on the nonlinear mapping of the network. To reduce computational complexity and to response promptly, BOW-LM employs a matrix correction technique to update the learning model by only $\mathcal {O}(L^2)$O(L2) operations for each data chunk with $L$L hidden layer nodes. Empirical studies on both synthetic and real-world datasets demonstrate the effectiveness of BOW-LM.

One-class collaborative filtering (OCCF) is a fundamental research problem in a myriad of applications where the preferences of users can only be implicitly inferred from their one-class feedback (e.g., click an ad or purchase a product). The main challenges of OCCF lie in the sparsity of user feedback and the ambiguity of unobserved preferences. To effectively address the above two challenges, side networks from users and items are extensively exploited by state-of-the-art methods, which are predominantly focused on static settings. However, as real-world recommender systems evolve over time (where both the user–item ratings and user–user/item–item side networks will change), it is necessary to update OCCF results (e.g., the latent features of users and items) accordingly. The main obstacle for OCCF online update with co-evolving side networks lies in the fact that the coupled system is highly sensitive to local changes, which may cause massive perturbation on the latent features of a large number of users and items. In this paper, we propose a novel incremental one-class collaborative filtering (OCCF) method that can cope with co-evolving side networks efficiently. In particular, we model the evolution of latent features as a linear transformation process, which enables fast update of the latent features on the fly. Empirical experiments demonstrate that our method can provide high-quality recommendation results on real-world datasets.

Knowledge graphs (KGs) are of great importance in various artificial intelligence systems, such as question answering, relation extraction, and recommendation. Nevertheless, most real-world KGs are highly incomplete, with many missing relations between entities. To discover new triples (i.e., head entity, relation, tail entity), many KG completion algorithms have been proposed in recent years. However, a vast majority of existing studies often require a large number of training triples for each relation, which contradicts the fact that the frequency distribution of relations in KGs often follows a long tail distribution, meaning a majority of relations have only very few triples. Meanwhile, since most existing large-scale KGs are constructed automatically by extracting information from crowd-sourcing data using heuristic algorithms, plenty of errors could be inevitably incorporated due to the lack of human verification, which greatly reduces the performance for KG completion. To tackle the aforementioned issues, in this paper, we study a novel problem of error-aware few-shot KG completion and present a principled KG completion framework REFORM. Specifically, we formulate the problem under the few-shot learning framework, and our goal is to accumulate meta-knowledge across different meta-tasks and generalize the accumulated knowledge to the meta-test task for error-aware few-shot KG completion. To address the associated challenges resulting from insufficient training samples and inevitable errors, we propose three essential modules neighbor encoder, cross-relation aggregation, and error mitigation in each meta-task. Extensive experiments on three widely used KG datasets demonstrate the superiority of the proposed framework REFORM over competitive baseline methods.

Feature selection is a prevalent data preprocessing paradigm for various learning tasks. Due to the expensive cost of acquiring supervision information, unsupervised feature selection sparks great interests recently. However, existing unsupervised feature selection algorithms do not have fairness considerations and suffer from a high risk of amplifying discrimination by selecting features that are over associated with protected attributes such as gender, race, and ethnicity. In this paper, we make an initial investigation of the fairness-aware unsupervised feature selection problem and develop a principled framework, which leverages kernel alignment to find a subset of high-quality features that can best preserve the information in the original feature space while being minimally correlated with protected attributes. Specifically, different from the mainstream in-processing debiasing methods, our proposed framework can be regarded as a model-agnostic debiasing strategy that eliminates biases and discrimination before downstream learning algorithms are involved. Experimental results on multiple real-world datasets demonstrate that our framework achieves a good trade-off between utility maximization and fairness promotion.

Prevailing methods of graph representation learning (GRL) usually rely on learning from "big'' data, requiring a large amount of labeled data for model training. However, it is common that graphs are associated with "small'' labeled data as data annotation and labeling is always a time and resource consuming task. The fact overshadows GRL's capability and applicability for many real situations. Therefore, data efficient learning on graphs has become essential for many real-world applications and there have been many studies working on this topic in recent years. In this tutorial, we will systematically review recent studies of data efficient learning on graphs, in particular a series of methods and applications of graph few-shot learning and graph self-supervised learning. At first, we will introduce the overview of graph representation learning methods, conventional few-shot learning, and self-supervised learning techniques. Then, we will present the work of data efficient learning on graphs in terms of three major graph mining tasks at different granularity levels: node-level learning tasks, graph-level learning tasks, and edge-level learning tasks. In the end, we will conclude the tutorial and raise open problems and pressing issues in future research. The authors of this tutorial are active and productive researchers in this research area.

Social relations are often used to improve recommendation quality when user-item interaction data is sparse in recommender systems. Most existing social recommendation models exploit pairwise relations to mine potential user preferences. However, real-life interactions among users are very complicated and user relations can be high-order. Hypergraph provides a natural way to model complex high-order relations, while its potentials for improving social recommendation are under-explored. In this paper, we fill this gap and propose a multi-channel hypergraph convolutional network to enhance social recommendation by leveraging high-order user relations. Technically, each channel in the network encodes a hypergraph that depicts a common high-order user relation pattern via hypergraph convolution. By aggregating the embeddings learned through multiple channels, we obtain comprehensive user representations to generate recommendation results. However, the aggregation operation might also obscure the inherent characteristics of different types of high-order connectivity information. To compensate for the aggregating loss, we innovatively integrate self-supervised learning into the training of the hypergraph convolutional network to regain the connectivity information with hierarchical mutual information maximization. The experimental results on multiple real-world datasets show that the proposed model outperforms the SOTA methods, and the ablation study verifies the effectiveness of the multi-channel setting and the self-supervised task. The implementation of our model is available via https://github.com/Coder-Yu/RecQ., Comment: 11 pages, Accepted by WWW'21. Correct some typos in the previous version

One fundamental problem in causal inference is to learn the individual treatment effects (ITE) -- assessing the causal effects of a certain treatment (e.g., prescription of medicine) on an important outcome (e.g., cure of a disease) for each data instance, but the effectiveness of most existing methods is often limited due to the existence of hidden confounders. Recent studies have shown that the auxiliary relational information among data can be utilized to mitigate the confounding bias. However, these works assume that the observational data and the relations among them are static, while in reality, both of them will continuously evolve over time and we refer such data as time-evolving networked observational data. In this paper, we make an initial investigation of ITE estimation on such data. The problem remains difficult due to the following challenges: (1) modeling the evolution patterns of time-evolving networked observational data; (2) controlling the hidden confounders with current data and historical information; (3) alleviating the discrepancy between the control group and the treated group. To tackle these challenges, we propose a novel ITE estimation framework Dynamic Networked Observational Data Deconfounder (\mymodel) which aims to learn representations of hidden confounders over time by leveraging both current networked observational data and historical information. Additionally, a novel adversarial learning based representation balancing method is incorporated toward unbiased ITE estimation. Extensive experiments validate the superiority of our framework when measured against state-of-the-art baselines. The implementation can be accessed in \hrefhttps://github.com/jma712/DNDC https://github.com/jma712/DNDC.

Twitter has become a major social media platform since its launching in 2006, while complaints about bot accounts have increased recently. Although extensive research efforts have been made, the state-of-the-art bot detection methods fall short of generalizability and adaptability. Specifically, previous bot detectors leverage only a small fraction of user information and are often trained on datasets that only cover few types of bots. As a result, they fail to generalize to real-world scenarios on the Twittersphere where different types of bots co-exist. Additionally, bots in Twitter are constantly evolving to evade detection. Previous efforts, although effective once in their context, fail to adapt to new generations of Twitter bots. To address the two challenges of Twitter bot detection, we propose SATAR, a self-supervised representation learning framework of Twitter users, and apply it to the task of bot detection. In particular, SATAR generalizes by jointly leveraging the semantics, property and neighborhood information of a specific user. Meanwhile, SATAR adapts by pre-training on a massive number of self-supervised users and fine-tuning on detailed bot detection scenarios. Extensive experiments demonstrate that SATAR outperforms competitive baselines on different bot detection datasets of varying information completeness and collection time. SATAR is also proved to generalize in real-world scenarios and adapt to evolving generations of social media bots., Comment: accepted at CIKM 2021

With the prevalence of social media, there has recently been a proliferation of recommenders that shift their focus from individual modeling to group recommendation. Since the group preference is a mixture of various predilections from group members, the fundamental challenge of group recommendation is to model the correlations among members. Existing methods mostly adopt heuristic or attention-based preference aggregation strategies to synthesize group preferences. However, these models mainly focus on the pairwise connections of users and ignore the complex high-order interactions within and beyond groups. Besides, group recommendation suffers seriously from the problem of data sparsity due to severely sparse group-item interactions. In this paper, we propose a self-supervised hypergraph learning framework for group recommendation to achieve two goals: (1) capturing the intra- and inter-group interactions among users; (2) alleviating the data sparsity issue with the raw data itself. Technically, for (1), a hierarchical hypergraph convolutional network based on the user- and group-level hypergraphs is developed to model the complex tuplewise correlations among users within and beyond groups. For (2), we design a double-scale node dropout strategy to create self-supervision signals that can regularize user representations with different granularities against the sparsity issue. The experimental analysis on multiple benchmark datasets demonstrates the superiority of the proposed model and also elucidates the rationality of the hypergraph modeling and the double-scale self-supervision., Comment: 11 pages, 6 figures, CIKM 2021

Due to the expensive cost of data annotation, few-shot learning has attracted increasing research interests in recent years. Various meta-learning approaches have been proposed to tackle this problem and have become the de facto practice. However, most of the existing approaches along this line mainly focus on image and text data in the Euclidean domain. However, in many real-world scenarios, a vast amount of data can be represented as attributed networks defined in the non-Euclidean domain, and the few-shot learning studies in such structured data have largely remained nascent. Although some recent studies have tried to combine meta-learning with graph neural networks to enable few-shot learning on attributed networks, they fail to account for the unique properties of attributed networks when creating diverse tasks in the meta-training phase---the feature distributions of different tasks could be quite different as instances (i.e., nodes) do not follow the data i.i.d. assumption on attributed networks. Hence, it may inevitably result in suboptimal performance in the meta-testing phase. To tackle the aforementioned problem, we propose a novel graph meta-learning framework--Attribute Matching Meta-learning Graph Neural Networks (AMM-GNN). Specifically, the proposed AMM-GNN leverages an attribute-level attention mechanism to capture the distinct information of each task and thus learns more effective transferable knowledge for meta-learning. We conduct extensive experiments on real-world datasets under a wide range of settings and the experimental results demonstrate the effectiveness of the proposed AMM-GNN framework.

Networked observational data presents new opportunities for learning individual causal effects, which plays an indispensable role in decision making. Such data poses the challenge of confounding bias. Previous work presents two desiderata to handle confounding bias. On the treatment group level, we aim to balance the distributions of confounder representations. On the individual level, it is desirable to capture patterns of hidden confounders that predict treatment assignments. Existing methods show the potential of utilizing network information to handle confounding bias, but they only try to satisfy one of the two desiderata. This is because the two desiderata seem to contradict each other. When the two distributions of confounder representations are highly overlapped, then we confront the undiscriminating problem between the treated and the controlled. In this work, we formulate the two desiderata as a minimax game. We propose IGNITE that learns representations of confounders from networked observational data, which is trained by a minimax game to achieve the two desiderata. Experiments verify the efficacy of IGNITE on two datasets under various settings.

Attributed networks nowadays are ubiquitous in a myriad of high-impact applications, such as social network analysis, financial fraud detection, and drug discovery. As a central analytical task on attributed networks, node classification has received much attention in the research community. In real-world attributed networks, a large portion of node classes only contain limited labeled instances, rendering a long-tail node class distribution. Existing node classification algorithms are unequipped to handle the \textit{few-shot} node classes. As a remedy, few-shot learning has attracted a surge of attention in the research community. Yet, few-shot node classification remains a challenging problem as we need to address the following questions: (i) How to extract meta-knowledge from an attributed network for few-shot node classification? (ii) How to identify the informativeness of each labeled instance for building a robust and effective model? To answer these questions, in this paper, we propose a graph meta-learning framework -- Graph Prototypical Networks (GPN). By constructing a pool of semi-supervised node classification tasks to mimic the real test environment, GPN is able to perform \textit{meta-learning} on an attributed network and derive a highly generalizable model for handling the target classification task. Extensive experiments demonstrate the superior capability of GPN in few-shot node classification., Accpeted by CIKM 2020

In this project, we will explore the range of data-related decisions made during public health emergencies like the ongoing COVID-19 pandemic and analyze the flow of information, data, and metadata within networks of such decisions. Data sharing is now considered a key component of addressing present, future, and even past public health emergencies, from local to global levels. Researchers, research institutions, journals and others have taken steps towards increasing the sharing of data around the ongoing COVID-19 pandemic and in preparation for future pandemics. We will quantify the effects of data flow modifications to identify parameter sets under which specific modes of sharing or withholding information have the largest effects on outbreak dynamics. For these high-impact parameter sets, we will then assess the current and past availability of corresponding data, metadata, and misinformation, and estimate the effects on outbreak mitigation and preparedness efforts.

Social recommender systems are expected to improve recommendation quality by incorporating social information when there is little user-item interaction data. However, recent reports from industry show that social recommender systems consistently fail in practice. According to the negative findings, the failure is attributed to: (1) A majority of users only have a very limited number of neighbors in social networks and can hardly benefit from social relations; (2) Social relations are noisy but they are indiscriminately used; (3) Social relations are assumed to be universally applicable to multiple scenarios while they are actually multi-faceted and show heterogeneous strengths in different scenarios. Most existing social recommendation models only consider the homophily in social networks and neglect these drawbacks. In this paper we propose a deep adversarial framework based on graph convolutional networks (GCN) to address these problems. Concretely, for (1) and (2), a GCN-based autoencoder is developed to augment the relation data by encoding high-order and complex connectivity patterns, and meanwhile is optimized subject to the constraint of reconstructing the social profile to guarantee the validity of the identified neighborhood. After obtaining enough purified social relations for each user, a GCN-based attentive social recommendation module is designed to address (3) by capturing the heterogeneous strengths of social relations. Finally, we adopt adversarial training to unify all the components by playing a Minimax game and ensure a coordinated effort to enhance recommendation performance. Extensive experiments on multiple open datasets demonstrate the superiority of our framework and the ablation study confirms the importance and effectiveness of each component., Accepted by TKDE

Networked data is commonly observed in many high-impact domains, ranging from social networks, collaboration platforms to biological systems. In such systems, the nodes are often associated with high dimensional features while remain connected to each other through pairwise interactions. Recently, various unsupervised feature selection methods have been developed to distill actionable insights from such data by finding a subset of relevant features that are highly correlated with the observed node connections. Although practically useful, those methods predominantly assume that the nodes on the network are organized in a flat structure, which is rarely the case in reality. In fact, the nodes in most, if not all, of the networks can be organized into a hierarchical structure. For example, in a collaboration network, researchers can be clustered into different research areas at the coarsest level and are further specified into different sub-areas at a finer level. Recent studies have shown that such hierarchical structure can help advance various learning problems including clustering and matrix completion. Motivated by the success, in this paper, we propose a novel unsupervised feature selection framework (HNFS) on networked data. HNFS can simultaneously learn the implicit hierarchical structure among the nodes and embed the hierarchical structure into the feature selection process. Empirical evaluations on various real-world datasets validate the superiority of our proposed framework.

Counterfactual evaluation of novel treatment assignment functions (e.g., advertising algorithms and recommender systems) is one of the most crucial causal inference problems for practitioners. Traditionally, randomized controlled trials (A/B tests) are performed to evaluate treatment assignment functions. However, such trials can be time-consuming, expensive, and even unethical in some cases. Therefore, offline counterfactual evaluation of treatment assignment functions becomes a pressing issue because a massive amount of observational data is available in today's big data era. Counterfactual evaluation requires handling the hidden confounders -- the unmeasured features which causally influence both the treatment assignment and the outcome. To deal with the hidden confounders, most of the existing methods rely on the assumption of no hidden confounders. However, this assumption can be untenable in the context of massive observational data. When such data comes with network information, the later can be potentially useful to correct hidden confounding bias. As such, we first formulate a novel problem, counterfactual evaluation of treatment assignment functions with networked observational data. Then, we investigate the following research questions: How can we utilize network information in counterfactual evaluation? Can network information improve the estimates in counterfactual evaluation? Toward answering these questions, first, we propose a novel framework, \emph{Counterfactual Network Evaluator} (CONE), which (1) learns partial representations of latent confounders under the supervision of observed treatments and outcomes; and (2) combines them for counterfactual evaluation. Then through extensive experiments, we corroborate the effectiveness of CONE. The results imply that incorporating network information mitigates hidden confounding bias in counterfactual evaluation.

In a conventional supervised learning paradigm, each data instance is associated with one single class label. Multilabel learning differs in the way that data instances may belong to multiple concepts simultaneously, which naturally appear in a variety of high impact domains, ranging from bioinformatics and information retrieval to multimedia analysis. It targets leveraging the multiple label information of data instances to build a predictive learning model that can classify unlabeled instances into one or multiple predefined target classes. In multilabel learning, even though each instance is associated with a rich set of class labels, the label information could be noisy and incomplete as the labeling process is both time consuming and labor expensive, leading to potential missing annotations or even erroneous annotations. The existence of noisy and missing labels could negatively affect the performance of underlying learning algorithms. More often than not, multilabeled data often has noisy, irrelevant, and redundant features of high dimensionality. The existence of these uninformative features may also deteriorate the predictive power of the learning model due to the curse of dimensionality. Feature selection, as an effective dimensionality reduction technique, has shown to be powerful in preparing high-dimensional data for numerous data mining and machine-learning tasks. However, a vast majority of existing multilabel feature selection algorithms either boil down to solving multiple single-labeled feature selection problems or directly make use of the imperfect labels to guide the selection of representative features. As a result, they may not be able to obtain discriminative features shared across multiple labels. In this article, to bridge the gap between a rich source of multilabel information and its blemish in practical usage, we propose a novel noise-resilient multilabel informed feature selection framework (MIFS) by exploiting the correlations among different labels. In particular, to reduce the negative effects of imperfect label information in obtaining label correlations, we decompose the multilabel information of data instances into a low-dimensional space and then employ the reduced label representation to guide the feature selection phase via a joint sparse regression framework. Empirical studies on both synthetic and real-world datasets demonstrate the effectiveness and efficiency of the proposed MIFS framework.

A variety of real-world applications such as complex industry process usually are lack of abundant training samples since the data acquiring process is time and labor consuming. Hence, it is important to utilize the limited training samples to build a sophisticated data-driven model, which may improve industry productivity. Recently, nonlinear learning models such as artificial neural networks and support vector machines have shown to be effective in modeling small-scale data by their strong modeling ability. However, these nonlinear learning models work as a black box and are often not human understandable and are difficult to be interpreted. In addition, in many applications, domain experts could provide us valuable expertise knowledge which may help further improve the modeling process. In this paper, we propose to integrate expertise knowledge to the nonlinear learning model to advance the data-driven modeling process in real-world applications. Experimental results on six benchmark datasets and a real-world industry application validate the effectiveness of the proposed model.

The proliferation of networked data in various disciplines motivates a surge of research interests on network or graph mining. Among them, node classification is a typical learning task that focuses on exploiting the node interactions to infer the missing labels of unlabeled nodes in the network. A vast majority of existing node classification algorithms overwhelmingly focus on static networks and they assume the whole network structure is readily available before performing learning algorithms. However, it is not the case in many real-world scenarios where new nodes and new links are continuously being added in the network. Considering the streaming nature of networks, we study how to perform online node classification on this kind of streaming networks (a.k.a. online learning on streaming networks). As the existence of noisy links may negatively affect the node classification performance, we first present an online network embedding algorithm to alleviate this problem by obtaining the embedding representation of new nodes on the fly. Then we feed the learned embedding representation into a novel online soft margin kernel learning algorithm to predict the node labels in a sequential manner. Theoretical analysis is presented to show the superiority of the proposed framework of online learning on streaming networks (OLSN). Extensive experiments on real-world networks further demonstrate the effectiveness and efficiency of the proposed OLSN framework.

Attributed networks are pervasive in numerous of high-impact domains. As opposed to conventional plain networks where only pairwise node dependencies are observed, both the network topology and node attribute information are readily available on attributed networks. More often than not, the nodal attributes are depicted in a high-dimensional feature space and are therefore notoriously difficult to tackle due to the curse of dimensionality. Additionally, features that are irrelevant to the network structure could hinder the discovery of actionable patterns from attributed networks. Hence, it is important to leverage feature selection to find a high-quality feature subset that is tightly correlated to the network structure. Few of the existing efforts either model the network structure at a macro-level by community analysis or directly make use of the binary relations. Consequently, they fail to exploit the finer-grained tie strength information for feature selection and may lead to suboptimal results. Motivated by the sociology findings, in this work, we investigate how to harness the tie strength information embedded on the network structure to facilitate the selection of relevant nodal attributes. Methodologically, we propose a principled unsupervised feature selection framework ADAPT to find informative features that can be used to regenerate the observed links and further characterize the adaptive neighborhood structure of the network. Meanwhile, an effective optimization algorithm for the proposed ADAPT framework is also presented. Extensive experimental studies on various real-world attributed networks validate the superiority of the proposed ADAPT framework.

Recently, there is a surge of research interests in finding anomalous nodes upon attributed networks. However, a vast majority of existing methods fail to capture the evolution of the networks properly, as they regard them as static. Meanwhile, they treat all the attributes and the instances equally, ignoring the existence of noisy. To tackle these problems, we propose a novel dynamic anomaly detection framework based on residual analysis, namely AMAD. It leverages the small smooth disturbance between time stamps to characterize the evolution of networks for incrementally update. Experiments conducted on several datasets show the superiority of AMAD in detecting anomalies.

Studying recommender systems with implicit feedback has become increasingly important. However, most existing works are designed in an offline setting while online recommendation is quite challenging due to the one-class nature of implicit feedback. In this paper, we propose an online collaborative filtering method for implicit feedback. We highlight three critical issues of existing works. First, when positive feedback arrives sequentially, if we treat all the other missing items for this given user as the negative samples, the mis-classified items will incur a large deviation since some items might appear as the positive feedback in the subsequent rounds. Second, the cost of missing a positive feedback should be much higher than that of having a false-positive. Third, the existing works usually assume that a fixed model is given prior to the learning task, which could result in poor performance if the chosen model is inappropriate. To address these issues, we propose a unified framework for Online Collaborative Filtering with Implicit Feedback (OCFIF). Motivated by the regret aversion, we propose a divestiture loss to heal the bias derived from the past mis-classified negative samples. Furthermore, we adopt cost-sensitive learning method to efficiently optimize the implicit MF model without imposing a heuristic weight restriction on missing data. By leveraging meta-learning, we dynamically explore a pool of multiple models to avoid the limitations of a single fixed model so as to remedy the drawback of manual/heuristic model selection. We also analyze the theoretical bounds of the proposed OCFIF method and conduct extensive experiments to evaluate its empirical performance on real-world datasets.

Anomaly detection is a fundamental problem in data mining field with many real-world applications. A vast majority of existing anomaly detection methods predominately focused on data collected from a single source. In real-world applications, instances often have multiple types of features, such as images (ID photos, finger prints) and texts (bank transaction histories, user online social media posts), resulting in the so-called multi-modal data. In this paper, we focus on identifying anomalies whose patterns are disparate across different modalities, i.e., cross-modal anomalies. Some of the data instances within a multi-modal context are often not anomalous when they are viewed separately in each individual modality, but contains inconsistent patterns when multiple sources are jointly considered. The existence of multi-modal data in many real-world scenarios brings both opportunities and challenges to the canonical task of anomaly detection. On the one hand, in multi-modal data, information of different modalities may complement each other in improving the detection performance. On the other hand, complicated distributions across different modalities call for a principled framework to characterize their inherent and complex correlations, which is often difficult to capture with conventional linear models. To this end, we propose a novel deep structured anomaly detection framework to identify the cross-modal anomalies embedded in the data. Experiments on real-world datasets demonstrate the effectiveness of the proposed framework comparing with the state-of-the-art., Comment: 8 pages, in Proceedings of the 2019 International Joint Conference on Neural Networks (IJCNN)

Objective Ovarian cancer (OC) is one of the most common types of cancer in women. Accurately prediction of benign ovarian tumors (BOT) and OC has important practical value. Methods Our dataset consists of 349 Chinese patients with 49 variables including demographics, blood routine test, general chemistry, and tumor markers. Machine learning Minimum Redundancy – Maximum Relevance (MRMR) feature selection method was applied on the 235 patients’ data (89 BOT and 146 OC) to select the most relevant features, with which a simple decision tree model was constructed. The model was tested on the rest of 114 patients (89 BOT and 25 OC). The results were compared with the predictions produced by using the risk of ovarian malignancy algorithm (ROMA) and logistic regression model. Results Eight notable features were selected by MRMR, among which two were identified as the top features by the decision tree model: human epididymis protein 4 (HE4) and carcinoembryonic antigen (CEA). Particularly, CEA is a valuable marker for OC prediction in patients with low HE4. The model also yields better prediction result than ROMA. Conclusion Machine learning approaches were able to accurately classify BOT and OC. Our goal is to derive a simple predictive model which also carries a good performance. Using our approach, we obtained a model that consists of just two biomarkers, HE4 and CEA. The model is simple to interpret and outperforms the existing OC prediction methods. It demonstrates that the machine learning approach has good potential in predictive modeling for the complex diseases.

The explosion of modeling complex systems using attributed networks boosts the research on anomaly detection in such networks, which can be applied in various high-impact domains. Many existing attempts, however, do not seriously tackle the inherent multi-view property in attribute space but concatenate multiple views into a single feature vector, which inevitably ignores the incompatibility between heterogeneous views caused by their own statistical properties. Actually, the distinct but complementary information brought by multi-view data promises the potential for more effective anomaly detection than the efforts only based on single-view data. Furthermore, the abnormal patterns naturally behave diversely in different views, which coincides with people's desire to discover specific abnormality according to their preferences for views (attributes). Most existing methods cannot adapt to people's requirements as they fail to consider the idiosyncrasy of user preferences. Therefore, we propose a multi-view framework Alarm to incorporate user preferences into anomaly detection and simultaneously tackle heterogeneous attribute characteristics through multiple graph encoders and a well-designed aggregator that supports self-learning and user-guided learning. Experiments on synthetic and real-world datasets, e.g., Disney, Books, and Enron, corroborate the improvement of Alarm in detection accuracy evaluated by the AUC metric and its effectiveness in supporting user-oriented anomaly detection.

Network embedding has been increasingly used in many network analytics applications to generate low-dimensional vector representations, so that many off-the-shelf models can be applied to solve a wide variety of data mining tasks. However, similar to many other machine learning methods, network embedding results remain hard to be understood by users. Each dimension in the embedding space usually does not have any specific meaning, thus it is difficult to comprehend how the embedding instances are distributed in the reconstructed space. In addition, heterogeneous content information may be incorporated into network embedding, so it is challenging to specify which source of information is effective in generating the embedding results. In this paper, we investigate the interpretation of network embedding, aiming to understand how instances are distributed in embedding space, as well as explore the factors that lead to the embedding results. We resort to the post-hoc interpretation scheme, so that our approach can be applied to different types of embedding methods. Specifically, the interpretation of network embedding is presented in the form of a taxonomy. Effective objectives and corresponding algorithms are developed towards building the taxonomy. We also design several metrics to evaluate interpretation results. Experiments on real-world datasets from different domains demonstrate that, by comparing with the state-of-the-art alternatives, our approach produces effective and meaningful interpretation to embedding results.

Attributed networks are pervasive in different domains, ranging from social networks, gene regulatory networks to financial transaction networks. This kind of rich network representation presents challenges for anomaly detection due to the heterogeneity of two data representations. A vast majority of existing algorithms assume certain properties of anomalies are given a prior. Since various types of anomalies in real-world attributed networks co-exist, the assumption that priori knowledge regarding anomalies is available does not hold. In this paper, we investigate the problem of anomaly detection in attributed networks generally from a residual analysis perspective, which has been shown to be effective in traditional anomaly detection problems. However, it is a non-trivial task in attributed networks as interactions among instances complicate the residual modeling process. Methodologically, we propose a learning framework to characterize the residuals of attribute information and its coherence with network information for anomaly detection. By learning and analyzing the residuals, we detect anomalies whose behaviors are singularly different from the majority. Experiments on real datasets show the effectiveness and generality of the proposed framework.

Attributed network embedding aims to seek low-dimensional vector representations for nodes in a network, such that original network topological structure and node attribute proximity can be preserved in the vectors. These learned representations have been demonstrated to be helpful in many learning tasks such as network clustering and link prediction. While existing algorithms follow an unsupervised manner, nodes in many real-world attributed networks are often associated with abundant label information, which is potentially valuable in seeking more effective joint vector representations. In this paper, we investigate how labels can be modeled and incorporated to improve attributed network embedding. This is a challenging task since label information could be noisy and incomplete. In addition, labels are completely distinct with the geometrical structure and node attributes. The bewildering combination of heterogeneous information makes the joint vector representation learning more difficult. To address these issues, we propose a novel Label informed Attributed Network Embedding (LANE) framework. It can smoothly incorporate label information into the attributed network embedding while preserving their correlations. Experiments on real-world datasets demonstrate that the proposed framework achieves significantly better performance compared with the state-of-the-art embedding algorithms.

The recent popularity of big data has brought immense quantities of high-dimensional data, which presents challenges to traditional data mining tasks due to curse of dimensionality. Feature selection has shown to be effective to prepare these high dimensional data for a variety of learning tasks. To provide easy access to feature selection algorithms, we provide an interactive feature selection tool FeatureMiner based on our recently released feature selection repository scikit-feature. FeatureMiner eases the process of performing feature selection for practitioners by providing an interactive user interface. Meanwhile, it also gives users some practical guidance in finding a suitable feature selection algorithm among many given a specific dataset. In this demonstration, we show (1) How to conduct data preprocessing after loading a dataset; (2) How to apply feature selection algorithms; (3) How to choose a suitable algorithm by visualized performance evaluation.

Rule-based classifier has shown its popularity in building many decision support systems such as medical diagnosis and financial fraud detection. One major advantage is that the models are human understandable and can be edited. Associative classifiers, as an extension of rule-based classifiers, use association rules to associate attributes with class labels. A delicate issue of associative classifiers is the need for subtle thresholds: minimum support and minimum confidence. Without prior knowledge, it could be difficult to choose the proper thresholds, and the discovered rules within the support-confidence framework are not statistically significant, i.e., inclusion of noisy rules and exclusion of valuable rules. Besides, most associative classifiers proposed so far, are built with only positive association rules. Negative rules, however, are also able to provide valuable information to discriminate between classes. To solve the above mentioned problems, we propose a novel associative classifier which is built upon both positive and negative classification association rules that show statistically significant dependencies. Experimental results on real-world datasets show that our method achieves competitive or even better performance than well-known rule-based and associative classifiers in terms of both classification accuracy and computational efficiency.

The explosive growth of social media sites brings about massive amounts of high-dimensional data. Feature selection is effective in preparing high-dimensional data for data analytics. The characteristics of social media present novel challenges for feature selection. First, social media data is not fully structured and its features are usually not predefined, but are generated dynamically. For example, in Twitter, slang words (features) are created everyday and quickly become popular within a short period of time. It is hard to directly apply traditional batch-mode feature selection methods to find such features. Second, given the nature of social media, label information is costly to collect. It exacerbates the problem of feature selection without knowing feature relevance. On the other hand, opportunities are also unequivocally present with additional data sources; for example, link information is ubiquitous in social media and could be helpful in selecting relevant features. In this paper, we study a novel problem to conduct unsupervised streaming feature selection for social media data. We investigate how to exploit link information in streaming feature selection, resulting in a novel unsupervised streaming feature selection framework USFS. Experimental results on two real-world social media datasets show the effectiveness and efficiency of the proposed framework comparing with the state-of-the-art unsupervised feature selection algorithms.

The semi-supervised, density-based clustering algorithm SSDBSCAN extracts clusters of a given dataset from different density levels by using a small set of labeled objects. A critical assumption of SSDBSCAN is, however, that at least one labeled object for each natural cluster in the dataset is provided. This assumption may be unrealistic when only a very few labeled objects can be provided, for instance due to the cost associated with determining the class label of an object. In this paper, we introduce a novel active learning strategy to select “most representative” objects whose class label should be determined as input for SSDBSCAN. By incorporating a Laplacian Graph Regularizer into a Local Linear Reconstruction method, our proposed algorithm selects objects that can represent the whole data space well. Experiments on synthetic and real datasets show that using the proposed active learning strategy, SSDBSCAN is able to extract more meaningful clusters even when only very few labeled objects are provided.

Mining association rules associates events that took place together. In market basket analysis, these discovered rules associate items purchased together. Items that are not part of a transaction are not considered. In other words, typical association rules do not take into account items that are part of the domain but that are not together part of a transaction. Association rules are based on frequencies and count the transactions where items occur together. However, counting absences of items is prohibitive if the number of possible items is very large, which is typically the case. Nonetheless, knowing the relationship between the absence of an item and the presence of another can be very important in some applications. These rules are called negative association rules. We review current approaches for mining negative association rules and we discuss limitations and future research directions.

Co-location rule mining is one of the tasks of spatial data mining, which focuses on the detection of sets of spatial features that show spatial associations. Most previous methods are generally based on transaction-free apriori-like algorithms which are dependent on user-defined thresholds and are designed for boolean data points. Due to the absence of a clear notion of transactions, it is nontrivial to use association rule mining techniques to tackle the co-location rule mining problem. To solve these difficulties, a transactionization approach was recently proposed; designed to mine datasets with extended spatial objects. A statistical test is used instead of global thresholds to detect significant co-location rules. One major shortcoming of this work is that it limits the size of antecedent of co-location rules up to three features, therefore, the algorithm is difficult to scale up. In this paper we introduce a new algorithm that fully exploits the property of statistical significance to detect more general co-location rules. We use our algorithm on real datasets with the National Pollutant Release Inventory (NPRI). A classifier is also proposed to help evaluate the discovered co-location rules.