Dynamic hashtag recommendation in social media with trend shift detection and adaptation

2025-03-30

AI Summary

Static hashtag recommendation models struggle to handle dynamic tag emergence and semantic drift. To address this, we propose a dynamic hashtag recommendation method tailored for social media streaming data. Methodologically, we design a trend-aware, lightweight drift detection mechanism enabling online model adaptation with minimal new posts; we further develop an end-to-end streaming recommendation framework integrating BERT-based semantic modeling and Apache Storm–enabled real-time processing, augmented by a trend-sensitive dynamic fine-tuning strategy. Our key contribution lies in the first joint modeling of fine-grained semantic drift awareness and low-overhead online adaptation. Evaluated on real-world COVID-19 and 2020 U.S. election datasets, our approach achieves a 23.6% accuracy gain over static baselines, with end-to-end response latency under 800 ms—demonstrating substantial improvements in timeliness and robustness.

Technology Category

• Data Mining & Knowledge Management: Data Stream Mining
• Machine Learning: Time-Series/Data Streams
• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

The widespread use of social media platforms results in the generation of vast amounts of user-generated content, which requires efficient methods for categorization and search. Hashtag recommendation systems have emerged as a crucial tool for automatically suggesting relevant hashtags and improving content discoverability. However, existing static models struggle to adapt to the highly dynamic and real-time nature of social media conversations, where new hashtags emerge and existing ones undergo semantic shifts. To address these challenges, this paper presents H-ADAPTS (Hashtag recommendAtion by Detecting and adAPting to Trend Shifts), a BERT-based hashtag recommendation methodology that can detect and adapt to shifts in the main trends and topics underlying social media conversation. Our approach introduces a trend-aware detection mechanism to identify changes in hashtag usage, triggering efficient model adaptation on a (small) set of recent posts. The framework leverages Apache Storm for real-time stream processing, enabling scalable and fault-tolerant analysis of high-velocity social data. Experimental results on two real-world case studies, including the COVID-19 pandemic and the 2020 US presidential election, demonstrate the ability to maintain high recommendation accuracy by adapting to emerging trends. Our methodology significantly outperforms existing solutions, ensuring timely and relevant hashtag recommendations in dynamic environments.

Problem

• Dynamic hashtag recommendation for real-time social media trends
• Detecting and adapting to semantic shifts in hashtag usage
• Improving content discoverability with scalable, trend-aware models

Innovation

• BERT-based dynamic hashtag recommendation system
• Trend-aware detection for semantic shift adaptation
• Apache Storm for real-time stream processing