CobwebTM: Probabilistic Concept Formation for Lifelong and Hierarchical Topic Modeling
For researchers in topic modeling and lifelong learning, CobwebTM offers a low-parameter, adaptive alternative to neural and classical models, addressing catastrophic forgetting and fixed capacity.
CobwebTM introduces a lifelong hierarchical topic model that combines incremental probabilistic concept formation with continuous document embeddings, achieving strong topic coherence and stable topics without predefining topic numbers. It demonstrates efficient unsupervised topic discovery across diverse datasets.
Topic modeling seeks to uncover latent semantic structure in text corpora with minimal supervision. Neural approaches achieve strong performance but require extensive tuning and struggle with lifelong learning due to catastrophic forgetting and fixed capacity, while classical probabilistic models lack flexibility and adaptability to streaming data. We introduce CobwebTM, a low-parameter lifelong hierarchical topic model based on incremental probabilistic concept formation. By adapting the Cobweb algorithm to continuous document embeddings, CobwebTM constructs semantic hierarchies online, enabling unsupervised topic discovery, dynamic topic creation, and hierarchical organization without predefining the number of topics. Across diverse datasets, CobwebTM achieves strong topic coherence, stable topics over time, and high-quality hierarchies, demonstrating that incremental symbolic concept formation combined with pretrained representations is an efficient approach to topic modeling.