ACM Transactions on Knowledge Discovery from Data最新文献

The fastest route search, which is to find a path with the shortest travel time when the user initiates a query, has become one of the most important services in many map applications. To enhance the user experience of travel, it is necessary to achieve accurate and real-time route search. However, traffic conditions are changing dynamically, especially the frequent occurrence of traffic congestion may greatly increase travel time. Thus, it is challenging to achieve the above goal. To deal with it, we present a congestion-aware spatio-temporal graph convolutional network based A* search algorithm for the task of fastest route search. We first identify a sequence of consecutive congested traffic conditions as a traffic congestion event. Then, we propose a spatio-temporal graph convolutional network that jointly models the congestion events and changing travel time to capture their complex spatio-temporal correlations, which can predict the future travel time information of each road segment as the basis of route planning. Further, we design a path-aided neural network to achieve effective origin-destination (OD) shortest travel time estimation by encoding the complex relationships between OD pairs and their corresponding fastest paths. Finally, the cost function in the A* algorithm is set by fusing the output results of the two components, which is used to guide the route search. Our experimental results on the two real-world datasets show the superior performance of the proposed method.

In the financial sphere, there is a wealth of accumulated unstructured financial data, such as the textual disclosure documents that companies submit on a regular basis to regulatory agencies, such as the Securities and Exchange Commission (SEC). These documents are typically very long and tend to contain valuable soft information about a company’s performance that is not present in quantitative predictors. It is therefore of great interest to learn predictive models from these long textual documents, especially for forecasting numerical key performance indicators (KPIs). In recent years, there has been a great progress in natural language processing via pre-trained language models (LMs) learned from large corpora of textual data. This prompts the important question of whether they can be used effectively to produce representations for long documents, as well as how we can evaluate the effectiveness of representations produced by various LMs. Our work focuses on answering this critical question, namely the evaluation of the efficacy of various LMs in extracting useful soft information from long textual documents for prediction tasks. In this paper, we propose and implement a deep learning evaluation framework that utilizes a sequential chunking approach combined with an attention mechanism. We perform an extensive set of experiments on a collection of 10-K reports submitted annually by US banks, and another dataset of reports submitted by US companies, in order to investigate thoroughly the performance of different types of language models. Overall, our framework using LMs outperforms strong baseline methods for textual modeling as well as for numerical regression. Our work provides better insights into how utilizing pre-trained domain-specific and fine-tuned long-input LMs for representing long documents can improve the quality of representation of textual data, and therefore, help in improving predictive analyses.

Federated Learning (FL) provided a novel paradigm for privacy-preserving machine learning, enabling multiple clients to collaborate on model training without sharing private data. To handle multi-source heterogeneous data, vertical federated learning (VFL) has been extensively investigated. However, in the context of VFL, the label information tends to be kept in one authoritative client and is very limited. This poses two challenges for model training in the VFL scenario: On the one hand, a small number of labels cannot guarantee to train a well VFL model with informative network parameters, resulting in unclear boundaries for classification decisions; On the other hand, the large amount of unlabeled data is dominant and should not be discounted, and it’s worthwhile to focus on how to leverage them to improve representation modeling capabilities. In order to address the above two challenges, Firstly, we introduce supervised contrastive loss to enhance the intra-class aggregation and inter-class estrangement, which is to deeply explore label information and improve the effectiveness of downstream classification tasks. Secondly, for unlabeled data, we introduce a pseudo-label-guided consistency mechanism to induce the classification results coherent across clients, which allows the representations learned by local networks to absorb the knowledge from other clients, and alleviates the disagreement between different clients for classification tasks. We conduct sufficient experiments on four commonly used datasets, and the experimental results demonstrate that our method is superior to the state-of-the-art methods, especially in the low-label rate scenario, and the improvement becomes more significant.

In this paper, we study the problem of computing Random Forest-distances in the presence of missing data. We present a general framework which avoids pre-imputation and uses in an agnostic way the information contained in the input points. We centre our investigation on RatioRF, an RF-based distance recently introduced in the context of clustering and shown to outperform most known RF-based distance measures. We also show that the same framework can be applied to several other state-of-the-art RF-based measures and provide their extensions to the missing data case. We provide significant empirical evidence of the effectiveness of the proposed framework, showing extensive experiments with RatioRF on 15 datasets. Finally, we also positively compare our method with many alternative literature distances, which can be computed with missing values.

Outlier detection stands as a critical cornerstone in the field of data mining, with a wide range of applications spanning from fraud detection to network security. However, real-world scenarios often lack labeled data for training, necessitating unsupervised outlier detection methods. This study centers on Unsupervised Outlier Model Selection (UOMS), with a specific focus on the family of Internal, Relative Evaluation of Outlier Solutions (IREOS) algorithms. IREOS measures outlier candidate separability by evaluating multiple maximum-margin classifiers and, while effective, it is constrained by its high computational demands. We investigate the impact of several different separation methods in UOMS in terms of ranking quality and runtime. Surprisingly, our findings indicate that different separability measures have minimal impact on IREOS’ effectiveness. However, using linear separation methods within IREOS significantly reduces its computation time. These insights hold significance for real-world applications where efficient outlier detection is critical. In the context of this work, we provide the code for the IREOS algorithm and our separability techniques.

Social relations are often used as auxiliary information to address data sparsity and cold-start issues in social recommendations. In the real world, social relations among users are complex and diverse. Widely used graph neural networks (GNNs) can only model pairwise node relationships and are not conducive to exploring higher-order connectivity, while hypergraph provides a natural way to model high-order relations between nodes. However, recent studies show that social recommendations still face the following challenges: 1) a majority of social recommendations ignore the impact of multifaceted social relationships on user preferences; 2) the item homogeneity is often neglected, mainly referring to items with similar static attributes have similar attractiveness when exposed to users that indicating hidden links between items; and 3) directly combining the representations learned from different independent views cannot fully exploit the potential connections between different views. To address these challenges, in this paper, we propose a novel method DH-HGCN++ for multiple social recommendations. Specifically, dual homogeneity (i.e., social homogeneity and item homogeneity) is introduced to mine the impact of diverse social relations on user preferences and enrich item representations. Hypergraph convolution networks with motifs are further exploited to model the high-order relations between nodes. Finally, cross-view contrastive learning is proposed as an auxiliary task to jointly optimize the DH-HGCN++. Real-world datasets are used to validate the effectiveness of the proposed model, where we use sentiment analysis to extract comment relations and employ the k-means clustering algorithm to construct the item-item correlation graph. Experiment results demonstrate that our proposed method consistently outperforms the state-of-the-art baselines on Top-N recommendations.

Static analysis tools for capturing bugs and vulnerabilities in software programs are widely employed in practice, as they have the unique advantages of high coverage and independence from the execution environment. However, existing tools for analyzing large codebases often produce a great deal of false warnings over genuine bug reports. As a result, developers are required to manually inspect and confirm each warning, a challenging, time-consuming, and automation-essential task.

This paper advocates a fast, general, and easily extensible approach called Llm4sa that automatically inspects a sheer volume of static warnings by harnessing (some of) the powers of Large Language Models (LLMs). Our key insight is that LLMs have advanced program understanding capabilities, enabling them to effectively act as human experts in conducting manual inspections on bug warnings with their relevant code snippets. In this spirit, we propose a static analysis to effectively extract the relevant code snippets via program dependence traversal guided by the bug warnings reports themselves. Then, by formulating customized questions that are enriched with domain knowledge and representative cases to query LLMs, Llm4sa can remove a great deal of false warnings and facilitate bug discovery significantly. Our experiments demonstrate that Llm4sa is practical in automatically inspecting thousands of static warnings from Juliet benchmark programs and 11 real-world C/C++ projects, showcasing a high precision (81.13%) and a recall rate (94.64%) for a total of 9,547 bug warnings. Our research introduces new opportunities and methodologies for using the LLMs to reduce human labor costs, improve the precision of static analyzers, and ensure software trustworthiness.

Anomaly detection for multivariate time series is an essential task in the modern industrial field. Although several methods have been developed for anomaly detection, they usually fail to effectively exploit the metrical-temporal correlation and the other dependencies among multiple variables. To address this problem, we propose a stacked attention autoencoder for anomaly detection in multivariate time series (SA2E-AD); it focuses on fully utilizing the metrical and temporal relationships among multivariate time series. We design a multiattention block, alternately containing the temporal attention and metrical attention components in a hierarchical structure to better reconstruct normal time series, which is helpful in distinguishing the anomalies from the normal time series. Meanwhile, a two-stage training strategy is designed to further separate the anomalies from the normal data. Experiments on three publicly available datasets show that SA2E-AD outperforms the advanced baseline methods in detection performance and demonstrate the effectiveness of each part of the process in our method.

Existing methods based on transfer learning leverage auxiliary information to help tail generalization and improve the performance of the tail classes. However, they cannot fully exploit the relationships between auxiliary information and tail classes and bring irrelevant knowledge to the tail classes. To solve this problem, we propose a hierarchical CNN with knowledge complementation, which regards hierarchical relationships as auxiliary information and transfers relevant knowledge to tail classes. First, we integrate semantics and clustering relationships as hierarchical knowledge into the CNN to guide feature learning. Then, we design a complementary strategy to jointly exploit the two types of knowledge, where semantic knowledge acts as a prior dependence and clustering knowledge reduces the negative information caused by excessive semantic dependence (i.e., semantic gaps). In this way, the CNN facilitates the utilization of the two complementary hierarchical relationships and transfers useful knowledge to tail data to improve long-tailed classification accuracy. Experimental results on public benchmarks show that the proposed model outperforms existing methods. In particular, our model improves accuracy by 3.46% compared with the second-best method on the long-tailed tieredImageNet dataset.

In recent years, to address the issue of networked data sparsity in node classification tasks, cross-network node classification (CNNC) leverages the richer information from a source network to enhance the performance of node classification in the target network, which typically has sparser information. However, in real-world applications, labeled nodes may be collected from multiple sources with multiple modalities (e.g., text, vision, and video). Naive application of single-source and single-modal CNNC methods may result in sub-optimal solutions. To this end, in this paper, we propose a model called M²CDNE (Multi-source and Multi-modal Cross-network Deep Network Embedding) for cross-network node classification. In M²CDNE, we propose a deep multi-modal network embedding approach that combines the extracted deep multi-modal features to make the node vector representations network-invariant. In addition, we apply dynamic adversarial adaptation to assess the significance of marginal and conditional probability distributions between each source and target network to make node vector representations label-discriminative. Furthermore, we devise to classify nodes in the target network through the related source classifier and aggregate different predictions utilizing respective network weights, corresponding to the discrepancy between each source and target network. Extensive experiments performed on real-world datasets demonstrate that the proposed M²CDNE significantly outperforms the state-of-the-art approaches.