Short text classification research encounters semantic bias issues stemming from data sparsity and imbalance in real-world applications. Traditional deep learning methodologies focus on acquiring information at the token and feature levels. However, the imbalance and complexity of semantic information in short text scenarios limit the effectiveness of feature representation. To address this issue, we propose the Sparse Self-Attention Network (SSAN) for short text classification, a novel model specifically designed for this task. This network integrates attention mechanisms from three perspectives: tokens, features, and local semantics. It overcomes the limitations of traditional attention mechanisms in capturing complex feature relationships and introduces an adaptive sparse strategy to automatically focus on relevant tokens and features while filtering out redundant information, thereby enhancing the quality of feature extraction. The implementation steps include deep standardization and enhancement through the feature enhancement module, detailed analysis of the text using a multi-perspective self-attention mechanism, and, finally, optimizing feature representation with an adaptive sparse strategy. Experimental results demonstrate that SSAN outperforms existing methods on multiple benchmark datasets, achieving a 2.43% increase in accuracy. This validates its effectiveness and practical value in enhancing short text classification performance.

With the popularization of social networks, fake news is also widely and rapidly spreading, which poses a great threat to the Internet. Therefore, how to detect fake news automatically and efficiently has become an urgent problem to be solved. However, the existing approaches mostly focus on the explicit features (images and text) and deep fusions, without considering potential features such as text emotion and image category. To find a solution to this issue, we propose a Potential Features Fusion Network (PFFN), which models the explicit and potential features at the same time. To exploit the potential image features, we introduce a mixture of experts structure to process the news image separately, which can best use the relationships between the news image category and fake news detection. Besides, we also extract emotion features as potential text features and fuse them with explicit text features. Finally, we establish an attention-based feature fusion network to fuse the potential features with the explicit features, which can obtain a multimodal fusion feature of a piece of news and thus further improve the performance. We make experiments on four public datasets (Weibo16, Weibo19, Twitter, and PolitiFact); the results compared with the baseline approaches demonstrate that our PFFN has a better performance. Our code is available at https://github.com/Wang-bupt/PFFN

Due to its low storage cost and fast search speed, cross-modal retrieval based on hashing has attracted widespread attention and is widely used in real-world applications of social media search. However, most existing hashing methods are often limited by uncomprehensive feature representations and semantic associations, which greatly restricts their performance and applicability in practical applications. To deal with this challenge, in this paper, we propose an end-to-end graph attention network hashing (EGATH) for cross-modal retrieval, which can not only capture direct semantic associations between images and texts but also match semantic content between different modalities. We adopt the contrastive language image pretraining (CLIP) combined with the Transformer to improve understanding and generalization ability in semantic consistency across different data modalities. The classifier based on graph attention network is applied to obtain predicted labels to enhance cross-modal feature representation. We construct hash codes using an optimization strategy and loss function to preserve the semantic information and compactness of the hash code. Comprehensive experiments on the NUS-WIDE, MIRFlickr25K, and MS-COCO benchmark datasets show that our EGATH significantly outperforms against several state-of-the-art methods.