Stock trend prediction, the task of forecasting future trends of stocks from their historical feature sequences, remains highly challenging due to the complex and dynamic nature of financial markets. In reality, stocks form diverse relationships that transcend traditional sector boundaries as market conditions evolve, i.e., stocks within the same sector may display different trends, while those in different sectors often exhibit similar movements. However, most existing stock prediction methods rely on predefined static relationships, lacking flexibility to adapt to changing market dynamics. Furthermore, objectives widely adopted in prior work have limitations in capturing complex patterns and relationships in stock market data. To address these limitations, we propose DYCOR, a novel stock trend prediction method that integrates two key innovations: (i) dynamic stock clustering, which captures market characteristics without relying on predefined relationship data by adaptively discovering hidden stock relationships; and (ii) correlation-aware training, which aligns predicted and ground-truth stock trends by reflecting their correlations in a fine-grained manner. We evaluate DYCOR on three datasets NASDAQ, NYSE, and S&P 500 widely used in existing research, and this method demonstrates superior performance across correlation-based and retrieval-based metrics compared to state-of-the-art baseline methods, while maintaining competitive runtime efficiency.

Beyond using multimodal features of items in addition to user-item interactions, researchers have additionally utilized Contrastive Learning (CL) in recent multimedia recommender systems to highly alleviate the data sparsity problem. CL-based methods generate at least two embeddings (i.e., views) for each instance and enrich the information of each instance from various perspectives via the views, thereby alleviating the data sparsity problem. Therefore, CL-based methods have focused on generating views that effectively represent the characteristics of each instance for their downstream tasks. Similarly, CL-based multimedia recommender systems have made efforts to effectively generate their user/item views by leveraging items' multimodal features. However, we point out the following two limitations that they have overlooked in generating their views: (1) they either have not attempted to identify the influence of each modality feature of an item on user-item interactions, or have identified it by randomly masking or dropping user-item interactions, and (2) they have not attempted to identify non-interactions likely to result in interactions in the future. To overcome these limitations, we propose a novel multimedia recommendation framework, named STARLINE, utilizing contraSTive leARning with modaLIty-aware graph refiNEment. Extensive experiments on five real-world datasets validate the effectiveness and validity of STARLINE, especially showing consistently higher accuracy by up to 13.24% compared to the best competitor.

The goal of Review-Based Recommendation System (RBRS) is to effectively learn the representations of users and items by utilizing review texts in addition to user-item interactions. From user-item interaction graphs widely employed in recommendation systems, recent RBRS methods using graph neural networks (GNNs) obtain the representations by associating each edge between a user and an item with the review information of the user for that item. However, these GNN-based RBRS methods present two main issues: (1) by con- verting each review text into the weight, i.e., single value, of a edge between a user node and an item node, they lose the rich informa- tion about users and items inherent in the review; and (2) by creating only a single general representation for each user, they cannot repre- sent the individual effects of users' likes and dislikes on their ratings for items they have interacted with. To address these problems, we propose a novel GNN-based RBRS, named LETTER, utilizing homo- geneous graphs, i.e., user-user graphs and an item-item graph, to learn general representations of users and items along with users' like and dislike representations. LETTER can learn user and item representations without losing review information by utilizing the proposed homogeneous graphs. Furthermore, LETTER explicitly designs the influence of users' like and dislike representations on their ratings to perform accurate rating predictions. Through ex- periments on six datasets, we verify that the proposed LETTER out- performs nine state-of-the-art RBRSs by up to 23.1%. Our source code is available at https://github.com/Bigdasgit/LETTER.

Fine-tuning pre-trained language models (PLMs) has recently shown a potential to improve knowledge graph completion (KGC). However, most PLM-based methods focus solely on encoding textual information, neglecting the long-tailed nature of knowledge graphs and their various topological structures, e.g., subgraphs, shortest paths, and degrees. We claim that this is a major obstacle to achieving higher accuracy of PLMs for KGC. To this end, we propose a Subgraph-Aware Training framework for KGC (SATKGC) with two ideas: (i) subgraph-aware mini-batching to encourage hard negative sampling and to mitigate an imbalance in the frequency of entity occurrences during training, and (ii) new contrastive learning to focus more on harder in-batch negative triples and harder positive triples in terms of the structural properties of the knowledge graph. To the best of our knowledge, this is the first study to comprehensively incorporate the structural inductive bias of the knowledge graph into fine-tuning PLMs. Extensive experiments on three KGC benchmarks demonstrate the superiority of SATKGC. Our code is available.https://github.com/meaningful96/SATKGC

A k -defective clique is a relaxation of the traditional clique definition, allowing up to k missing edges. This relaxation is crucial in various real-world applications such as link prediction, community detection, and social network analysis. Although the problems of enumerating maximal k -defective cliques and searching a maximum k -defective clique have been extensively studied, existing algorithms suffer from limitations such as the combinatorial explosion of small partial solutions and sub-optimal search spaces. To address these limitations, we propose a novel clique-first branch-and-bound framework that first generates cliques and then adds missing edges. Furthermore, we introduce a new pivoting technique that achieves a search space size of O (3 n/3 • n k ), where n is the number of vertices in the input graph. We prove that the worst-case number of maximal k -defective cliques is Ω(3 n/3 • n k ) when k is a constant, establishing that our algorithm's search space is worst-case optimal. Leveraging the diameter-two property of defective cliques, we further reduce the search space size to O (n • 3 δ/3 • (δ Δ) k ), where δ is the degeneracy and Δ is the maximum degree of the input graph. We also propose an efficient framework for maximum k -defective clique search based on our branch-and-bound, together with practical techniques to reduce the search space. Experiments on real-world benchmark datasets with more than 1 million edges demonstrate that each of our proposed algorithms for maximal k -defective clique enumeration and maximum k -defective clique search outperforms the respective state-of-the-art algorithms by up to four orders of magnitude in terms of processing time.

To recommend the points of interest (POIs) that a user would check-in next, most deep-learning (DL)-based existing studies have employed random negative (RN) sampling during model training. In this paper, we claim and validate that, as the training proceeds, such an RN sampling in reality performs as sampling easy negative (EN) POIs (i.e., EN sampling) that a user was highly unlikely to check-in at her check-in time point. Furthermore, we verify that EN sampling is more disadvantageous in improving the accuracy than sampling hard negative (HN) POIs (i.e., HN sampling) that a user was highly likely to check-in. To address this limitation, we present the novel concept of the Degree of Positiveness (DoP), which can be formulated by two factors: (i) the degree to which a POI has the characteristics preferred by a user; (ii) the geographical distance between a user and a POI. Then, we propose a new model-training scheme based on HN sampling by using DoP. Using real-world datasets (i.e., NYC, TKY, and Brightkite), we demonstrate that all the state-of-the-art models trained by our scheme showed dramatic improvements in accuracy by up to about 82.8%.

Subgraph matching is the problem of searching for all embeddings of a query graph in a data graph, and subgraph query processing (also known as subgraph search) is to find all the data graphs that contain a query graph as subgraphs. Extensive research has been done to develop practical solutions for both problems. However, the existing solutions still show limited query processing time due to a lot of unnecessary computations in search. In this paper, we focus on exploring as compact search space as possible by using three techniques: (1) pruning by bipartite matching, (2) pruning by failing sets with bipartite matching, and (3) cell-wide verification. We propose a new algorithm BICE, which combines these three techniques. We conduct extensive experiments on real-world datasets as well as synthetic datasets to evaluate the effectiveness of the techniques. Experiments show that our approach outperforms the fastest existing subgraph search algorithm by up to two orders of magnitude in terms of elapsed time to process a query. Our approach also outperforms state-of-the-art subgraph matching algorithms by up to two orders of magnitude.

Subgraph query processing (also known as subgraph search) and subgraph matching are fundamental graph problems in many application domains. A lot of efforts have been made to develop practical solutions for these problems. Despite the efforts, existing algorithms showed limited running time and scalability in dealing with large and/or many graphs. In this paper, we propose a new subgraph search algorithm using equivalences of vertices in order to reduce search space: (1) static equivalence of vertices in a query graph that leads to an efficient matching order of the vertices, and (2) dynamic equivalence of candidate vertices in a data graph, which enables us to capture and remove redundancies in search space. These techniques for subgraph search also lead to an improved algorithm for subgraph matching. Experiments show that our approach outperforms state-of-the-art subgraph search and subgraph matching algorithms by up to several orders of magnitude with respect to query processing time.

Supergraph search is one of fundamental graph query processing problems in many application domains. Given a query graph and a set of data graphs, supergraph search is to find all the data graphs contained in the query graph as subgraphs. In existing algorithms, index construction or filtering approaches are computationally expensive, and search methods can cause redundant computations. In this paper, we introduce four new concepts to address these limitations: (1) DAG integration, (2) dynamic programming between integrated DAG and graph, (3) active-first search, and (4) relevance-size order, which together lead to a much faster and scalable algorithm for supergraph search. Extensive experiments with real datasets show that our approach outperforms state-of-the-art algorithms by up to orders of magnitude in terms of indexing time and query processing time.

Subgraph matching (or subgraph isomorphism) is one of the fundamental problems in graph analysis. Extensive research has been done to develop practical solutions for subgraph matching. The state-of-the-art algorithms such as \textsfCFL-Match and \textsfTurbo\textsubscriptiso convert a query graph into a spanning tree for obtaining candidates for each query vertex and obtaining a good matching order with the spanning tree. However, by using the spanning tree instead of the original query graph, it could lead to lower pruning power and a sub-optimal matching order. Another limitation is that they perform redundant computation in search without utilizing the knowledge learned from past computation. In this paper, we introduce three novel concepts to address these inherent limitations: 1) dynamic programming between a directed acyclic graph (DAG) and a graph, 2) adaptive matching order with DAG ordering, and 3) pruning by failing sets, which together lead to a much faster algorithm \textsfDAF for subgraph matching. Extensive experiments with real datasets show that \textsfDAF outperforms the fastest existing solution by up to orders of magnitude in terms of recursive calls as well as in terms of the elapsed time.