Conversational recommender systems (CRSs) aim to provide personalized item recommendations along with explanations based on the conversations with users. While advancements in language models (LMs) have facilitated CRSs, limitations remain when LMs lack sufficient knowledge about item features that are essential for accurate recommendations and appropriate explanations. To alleviate this issue, retrieval-augmented language models (RALMs) have been introduced; however, they introduce a new challenge: the inclusion of less-relevant knowledge in retrieved passages. To address this limitation, we propose a novel CRS framework, MOCHA, which enhances RALMs through a multi-stage item/feature selection with Chain-of-Thought (CoT) reasoning. Specifically, MOCHA systematically identifies relevant knowledge by first selecting the item to recommend and then selecting its features to explain; each selection is performed via CoT reasoning. Experimental results on two public CRS datasets demonstrate that MOCHA significantly improves the recommendation accuracy, and provides informative and factually-correct explanations for the recommended items.

In e-commerce platforms, profit-aware recommender systems aim to improve the platform's profits while maintaining high overall accuracy by recommending items with high profits as top-ranked items. We explore two issues faced by existing model-based profit-aware approaches (i.e., MBAs) when training recommendation models for profit enhancement. First, existing MBAs tend to inaccurately infer the item ranking without considering the user's current preference for each item through their profit-based weighting scheme. Second, through the point-wise learning-to-rank (LTR), the model is optimized solely for the preference score of each item independently rather than being directly optimized for the overall ranking of items. To tackle these issues, we propose a novel MBA that involves three key steps: (S1) defining the Current Preference incorporated with Profit (i.e., CPP) for items; (S2) classifying items through CPP; and (S3) training the model by list-wise LTR based on CPP. Extensive experimental results using real-world platform datasets demonstrate that our approach improves accuracy by approximately 4% and profits by about 24% compared to the best-competing method.

Multi-criteria (MC) recommender systems, which utilize MC rating information for recommendation, are increasingly widespread in various e-commerce domains. However, the MC recommendation using training-based collaborative filtering, requiring consideration of multiple ratings compared to single-criterion counterparts, often poses practical challenges in achieving state-of-the-art performance along with scalable model training. To solve this problem, we propose CA-GF, a training-free MC recommendation method, which is built upon criteria-aware graph filtering for efficient yet accurate MC recommendations. Specifically, first, we construct an item-item similarity graph using an MC user-expansion graph. Next, we design CA-GF composed of the following key components, including 1) criterion-specific graph filtering where the optimal filter for each criterion is found using various types of polynomial low-pass filters and 2) criteria preference-infused aggregation where the smoothed signals from each criterion are aggregated. We demonstrate that CA-GF is (a) efficient: providing computational efficiency, offering an extremely fast runtime of less than 0.2 seconds even on the largest benchmark dataset, (b) accurate: outperforming benchmark MC recommendation methods, achieving substantial accuracy gains up to 24% compared to the best competitor, and (c) interpretable: providing interpretations for the contribution of each criterion to the model prediction based on visualizations.

With the prevalence of social media platforms, accurately identifying the same users across different networks through network alignment has become crucial. Existing methods often struggle due to sparse or absent user-identifiable information (node attributes), highlighting the need for augmenting node attributes. However, research on attribute augmentation remains largely under-explored. In this study, we aim to design augmented attributes that enhance network alignment by reflecting three key structural C haracteristics: (C1) global structural characteristic, reflects the global network structure; (C2) seed-based structural characteristic, leverages cross-network structural information associated with seed nodes; (C3) multi-aspect structural characteristic, employs diverse structural relationship measures. To this end, we propose a novel approach for designing trustworthy Augmented Seed-baSed and multI-aspect STructurAl iNformaTion (ASSISTANT) attributes. To enhance alignment performance, we also present a learning module that utilizes a gate mechanism to select the most effective measure dynamically. Extensive experiments across various datasets demonstrate the following: 1) Our network alignment framework, which includes a gate mechanism module, significantly outperforms state-of-the-art methods in alignment accuracy; 2) other state-of-the-art methods using ASSISTANT attributes as input substantially boosts their own alignment accuracy; and 3) using only ASSISTANT attributes without any training process also leads to effective alignment, showcasing their high trustworthiness.

Traffic speed prediction is a crucial task for optimizing navigation systems and reducing traffic congestion. Although there have been efforts to improve the accuracy of speed prediction by incorporating auxiliary features, such as traffic flow, weather, and time, types of auxiliary features are limited and their detailed relationships with speed have not been explored yet. In our study, we present the individual spatio-temporal (IST) dependencies on flow and speed, and characterize three types of IST-dependencies with the flow-to-flow, speed-to-speed, and flow-to-speed graphs. Then, we propose Auxiliary feature-aided Attention Network (ARIAN), a novel approach to judiciously learning the degrees of IST-dependencies with the three graphs and predicting the future speed by leveraging various auxiliary features. Through comprehensive experiments using 3 real-world datasets, we validate the superiority of ARIAN over 10 state-of-the-art methods and the effectiveness of each auxiliary feature and each dependency learner in ARIAN.

In this paper, we focus on multimedia recommender systems using graph convolutional networks (GCNs) where the multimodal features as well as user-item interactions are employed together. Our study aims to exploit multimodal features more effectively in order to accurately capture users' preferences for items. To this end, we point out following two limitations of existing GCN-based multimedia recommender systems: (L1) although multimodal features of interacted items by a user can reveal her preferences on items, existing methods utilize GCN designed to focus only on capturing collaborative signals, resulting in insufficient reflection of the multimodal features in the final user/item embeddings; (L2) although a user decides whether to prefer the target item by considering its multimodal features, existing methods represent her as only a single embedding regardless of the target item's multimodal features and then utilize her embedding to predict her preference for the target item. To address the above issues, we propose a novel multimedia recommender system, named MONET, composed of following two core ideas: modality-embracing GCN (MeGCN) and target-aware attention. Through extensive experiments using four real-world datasets, we demonstrate i) the significant superiority of MONET over seven state-of-the-art competitors (up to 30.32% higher accuracy in terms of recall@20, compared to the best competitor) and ii) the effectiveness of the two core ideas in MONET. All MONET codes are available at https://github.com/Kimyungi/MONET.

For high-quality conversational recommender systems (CRS), it is important to recommend the suitable items by capturing the items' features mentioned in the dialog and to explain the appropriate ones among the various features of the recommended item. We argue that the CRS model should be a domain-expert who is (1) knowledgeable about the relationships between items and their various features and (2) able to explain the recommended item with its features relevant to dialog context. To this end, we propose a novel framework, named as LATTE, to pre-train each core module in CRS (i.e., the recommendation and the conversation module) through abundant external data. For the recommendation module, we pre-train the recommendation module to comprehensively understand the relationships between items and their various features by leveraging both multi-reviews and a knowledge graph. For pre-training the conversation module, we create the synthetic dialogs, which contain responses providing the explanation relevant to the dialog context by using all the items' features and dialog templates. Through extensive experiments on two public CRS datasets, we demonstrate that LATTE exhibits (1) the effectiveness of each module in LATTE, (2) the superiority over 7 state-of-the art methods, and (3) the interpretations based on visualization.

The multi-criteria (MC) recommender system, which leverages MC rating information in a wide range of e-commerce areas, is ubiquitous nowadays. Surprisingly, although graph neural networks (GNNs) have been widely applied to develop various recommender systems due to GNN's high expressive capability in learning graph representations, it has been still unexplored how to design MC recommender systems with GNNs. In light of this, we make the first attempt towards designing a GNN-aided MC recommender system. Specifically, rather than straightforwardly adopting existing GNN-based recommendation methods, we devise a novel criteria preference-aware light graph convolution (CPA-LGC ) method, which is capable of precisely capturing the criteria preference of users as well as the collaborative signal in complex high-order connectivities. To this end, we first construct an MC expansion graph that transforms user-item MC ratings into an expanded bipartite graph to potentially learn from the collaborative signal in MC ratings. Next, to strengthen the capability of criteria preference awareness, CPA-LGC incorporates newly characterized embeddings, including user-specific criteria-preference embeddings and item-specific criterion embeddings, into our graph convolution model. Through comprehensive evaluations using four real-world datasets, we demonstrate (a) the superiority over benchmark MC recommendation methods and benchmark recommendation methods using GNNs with tremendous gains, (b) the effectiveness of core components in CPA-LGC, and (c) the computational efficiency.

The discovery of community structures in social networks has gained considerable attention as a fundamental problem for various network analysis tasks. However, due to privacy concerns or access restrictions, the network structure is often unknown, thereby rendering established community detection approaches ineffective without costly data acquisition. To tackle this challenge, we present META-CODE, a novel end-to-end solution for detecting overlapping communities in networks with unknown topology via exploratory learning aided by easy-to-collect node metadata. Specifically, META-CODE consists of three steps: 1) initial network inference, 2) node-level community-affiliation embedding based on graph neural networks (GNNs) trained by our new reconstruction loss, and 3) network exploration via community-affiliation-based node queries, where Steps 2 and 3 are performed iteratively. Experimental results demonstrate that META-CODE exhibits (a) superiority over benchmark methods for overlapping community detection, (b) the effectiveness of our training model, and (c) fast network exploration.

In a news recommender system, a user tends to click on a news article if she is interested in its topic understood by looking at its title. Such a behavior is possible since, when viewing the title, humans naturally think of the contextual meaning of each title word by leveraging their own background knowledge. Motivated by this, we propose a novel personalized news recommendation framework CAST (Context-aware Attention network with a Selection module for Title word representation), which is capable of enriching title words by leveraging body text that fully provides the whole content of a given article as the context. Through extensive experiments, we demonstrate (1) the effectiveness of core modules in CAST, (2) the superiority of CAST over 9 state-of-the-art news recommendation methods, and (3) the interpretability with CAST.

Network alignment (NA) is the task of discovering node correspondences across different networks. Although NA methods have achieved remarkable success in a myriad of scenarios, their satisfactory performance is not without prior anchor link information and/or node attributes, which may not always be available. In this paper, we propose Grad-Align+, a novel NA method using node attribute augmentation that is quite robust to the absence of such additional information. Grad-Align+ is built upon a recent state-of-the-art NA method, the so-called Grad-Align, that gradually discovers only a part of node pairs until all node pairs are found. Specifically, Grad-Align+ is composed of the following key components: 1) augmenting node attributes based on nodes' centrality measures, 2) calculating an embedding similarity matrix extracted from a graph neural network into which the augmented node attributes are fed, and 3) gradually discovering node pairs by calculating similarities between cross-network nodes with respect to the aligned cross-network neighbor-pair. Experimental results demonstrate that Grad-Align+ exhibits (a) superiority over benchmark NA methods, (b) empirical validation of our theoretical findings, and (c) the effectiveness of our attribute augmentation module.

The discovery of community structures in social networks has gained significant attention since it is a fundamental problem in understanding the networks’ topology and functions. However, most social network data are collected from partially observable networks with both missing nodes and edges . In this article, we address a new problem of detecting overlapping community structures in the context of such an incomplete network, where communities in the network are allowed to overlap since nodes belong to multiple communities at once. To solve this problem, we introduce KroMFac , a new framework that conducts community detection via regularized nonnegative matrix factorization (NMF) based on the Kronecker graph model. Specifically, from an inferred Kronecker generative parameter matrix, we first estimate the missing part of the network. As our major contribution to the proposed framework, to improve community detection accuracy, we then characterize and select influential nodes (which tend to have high degrees) by ranking, and add them to the existing graph. Finally, we uncover the community structures by solving the regularized NMF-aided optimization problem in terms of maximizing the likelihood of the underlying graph. Furthermore, adopting normalized mutual information (NMI), we empirically show superiority of our KroMFac approach over two baseline schemes by using both synthetic and real-world networks.

Relation prediction is a fundamental task in network analysis which aims to predict the relationship between two nodes. Thus, this differes from the traditional link prediction problem predicting whether a link exists between a pair of nodes, which can be viewed as a binary classification task. However, in the heterogeneous information network (HIN) which contains multiple types of nodes and multiple relations between nodes, the relation prediction task is more challenging. In addition, the HIN might have missing relation types on some edges and missing node types on some nodes, which makes the problem even harder.

In this paper, we present CR-Graph (community reinforcement on graphs), a novel method that helps existing algorithms to perform more-accurate community detection (CD). Toward this end, CR-Graph strengthens the community structure of a given original graph by adding non-existent predicted intra-community edges and deleting existing predicted inter-community edges. To design CR-Graph, we propose the following two strategies: (1) predicting intra-community and inter-community edges (i.e., the type of edges) and (2) determining the amount of edges to be added/deleted. To show the effectiveness of CR-Graph, we conduct extensive experiments with various CD algorithms on 7 synthetic and 4 real-world graphs. The results demonstrate that CR-Graph improves the accuracy of all underlying CD algorithms universally and consistently.

Density-based spatial clustering of applications with noise (DBSCAN) is the most commonly used density-based clustering algorithm, where it can discover multiple clusters with arbitrary shapes. DBSCAN works properly when the input data type is homogeneous, but the DBSCAN's approach may not be sufficient when the input dataset has textual heterogeneity (e.g., when we intend to find clusters from geo-tagged posts on social media relevant to a certain point-of-interest (POI)), thus leading to poor performance. In this paper, we present DBSTexC, a new density-based clustering algorithm using spatio--textual information on Twitter. We first define POI-relevant and POI-irrelevant tweets as the records that contain and do not contain a POI name or its coherent variations, respectively. By taking into account the fractions of POI-relevant and POI-irrelevant tweets, our DBSTexC algorithm shows a much higher clustering quality than the DBSCAN case in terms of the F1 score and its variants. DBSTexC can be thought of as a generalized version of DBSCAN due to the findings that it performs identically as DBSCAN when the inputs are homogeneous and far outperforms DBSCAN when the heterogeneous input data type is given.

This study analyzes friendships in online social networks involving geographic distance with a geo-referenced Twitter dataset, which provides the exact distance between corresponding users. We start by introducing a strong definition of "friend" on Twitter, requiring bidirectional communication. Next, by utilizing geo-tagged mentions delivered by users to determine their locations, we introduce a two-stage distance estimation algorithm. As our main contribution, our study provides the following newly-discovered friendship degree related to the issue of space: The number of friends according to distance follows a double power-law (i.e., a double Pareto law) distribution, indicating that the probability of befriending a particular Twitter user is significantly reduced beyond a certain geographic distance between users, termed the separation point. Our analysis provides much more fine-grained social ties in space, compared to the conventional results showing a homogeneous power-law with distance.