Topic Modeling Theories

A study repository pairing theoretical notes with runnable Jupyter notebooks for seven topic-modeling approaches: Latent Dirichlet Allocation, Dynamic Topic Model, Topics Over Time, Correlated Topic Model, Structural Topic Model, Biterm Topic Model, and BERTopic.

Each notebook is self-contained and moves from foundational LDA concepts to neural, embedding-based methods, so it doubles as a reference for the assumptions behind each model and a working code example, rather than only a description of its software interface.