คอลเล็กชัน (Collections)
พบ 3 รายการ
(ค้นหา: "transformer")
วารสารวิชาการ
ระดับนานาชาติ (SCOPUS) Q1
GCTMR: Graph Contrastive Transformer for Multibehavior Recommendation
Multibehavior recommendation aims to leverage diverse user interaction types—such as views, clicks, and purchases—to better understand user intent and enhance recommendation accuracy. However, effectively modeling these heterogeneous behaviors remains challenging due to the sparsity, noise, over-smooth embedding, and dynamic nature of interaction sequences. To address these issues, we propose GCTMR (Graph Contrastive Transformer for Multibehavior Recommendation), a framework that integrates graph-based contrastive learning with a behavior-aware Transformer architecture. Specifically, the methodological contribution lies in three key innovations: 1) an adaptive sparse attention mechanism employing α -entmax that dynamically filters out noisy or irrelevant interactions to preserve the distinctiveness of user behavior representations; 2) a frequency-aware multibehavior encoder using FFT/IFFT decomposition that disentangles stable long-term preferences from short-term behavioral shifts across different interaction types; and 3) a simplified focal loss function that emphasizes hard-to-predict instances, improving robustness under data sparsity and behavioral ambiguity. The empirical validation is exceptionally strong, demonstrating improving of 22.43% on Tmall and 18.10% on Beibei over state-of-the-art baselines in HR@10 metrics, with particularly remarkable performance under sparse data conditions where traditional methods struggle most. Comprehensive ablation studies confirm that each component contributes meaningfully to the overall performance, while extensive hyperparameter analyses demonstrate the framework’s stability and practical applicability.
บทความประชุมวิชาการ
ระดับนานาชาติ
GraphBERT4FakeNews: Graph-aware Bidirectional Transformer Learning for English Fake News Detection
The rapid dissemination of fake news through online news platforms and social media has become a major challenge, significantly influencing public opinion, political decision-making, and social stability. Recent fake news detection methods have achieved promising performance using deep contextual language models such as Bidirectional Encoder Representations from Transformers (BERT). However, most existing approaches process news articles independently and primarily focus on textual semantics while overlooking structural relationships among related news articles. Consequently, valuable graph-based information, including semantic similarity, shared entities, common publishers, and topical relationships, remains underutilized. This paper proposes GraphBERT4FakeNews, a graph transformer framework that jointly learns contextual semantic representations and graph structural information for English fake news detection. First, BERT generates contextualized embeddings from news articles. A news graph is then constructed by connecting semantically related articles according to contextual similarity. Subsequently, a GraphBERT encoder performs graph-aware self-attention to capture both local and long-range structural dependencies without relying on conventional graph convolution operations. Finally, a classification layer predicts whether a news article is genuine or fake. The proposed framework is evaluated on the publicly available FakeNewsNet benchmark, including the GossipCop and PolitiFact datasets. Experimental results demonstrate that GraphBERT4FakeNews effectively integrates semantic understanding with graph structural reasoning, leading to improved fake news detection performance compared with conventional language models and graph neural network approaches.
วารสารวิชาการ
ระดับนานาชาติ (SCOPUS) Q1
BiTG4Rec: Bidirectional Transformer Graphs for Sequential-Social Recommendation
Sequential-social recommendation systems are essential for understanding users’ evolving interests and predicting their future behaviors. While existing methods employing bidirectional graph modeling have shown promising results in capturing user interactions, they face significant challenges in effectively leveraging bidirectional information and addressing dynamic user preferences. To overcome these limitations, we propose Bidirectional Transformer Graphs for Sequential-Social Recommendation (BiTG4Rec), a novel framework that integrates both supervised and self-supervised learning to analyze dynamic user preferences at sequential and social levels through bidirectional graphs. Our approach comprises three core components: 1) a Bidirectional Dynamic Graph Convolutional Network (BiDGCN) that models long-term sequential preferences, 2) a Bidirectional Dynamic Graph Attention Network (BiDGAT) that captures short-term sequential preferences, and 3) a Bidirectional Hyperbolic Graph Contrastive Learning module (BiHGCL) that extracts social preferences. To unify these diverse signals, we introduce a TransformerGated mechanism that dynamically integrates long-term, short-term, and social preferences. Extensive experiments on multiple real-world datasets demonstrate that BiTG4Rec consistently outperforms state-of-the-art methods, validating its effectiveness for sequential-social recommendation tasks.