BERTrend: Neural Topic Modeling for Emerging Trends Detection
Abstract: Detecting and tracking emerging trends and weak signals in large, evolving text corpora is vital for applications such as monitoring scientific literature, managing brand reputation, and surveilling critical infrastructure. Existing solutions often fail to capture the nuanced context or dynamically track evolving patterns over time. BERTrend, a novel method, addresses these limitations using neural topic modeling in an online setting. It introduces a new metric to quantify topic popularity over time by considering both the number of documents and update frequency. This metric classifies topics as noise, weak, or strong signals, flagging emerging, rapidly growing topics for further investigation. Evaluations on two large real-world datasets demonstrate BERTrend's ability to accurately detect and track meaningful weak signals while filtering out noise, offering a comprehensive solution for monitoring emerging trends in large-scale, evolving text corpora.
Paper Type: Long
Research Area: Interpretability and Analysis of Models for NLP
Research Area Keywords: topic modeling, information extraction, zero/few-shot extraction, prompting, large language models, trend analysis
Contribution Types: Model analysis & interpretability, NLP engineering experiment, Theory
Languages Studied: English
Submission Number: 1476
Loading