Navigate a repository using topic modeling

Phaskell's platform  covers topics related to software development, technology trends, and coding tutorials. Through articles, tutorials, and personal anecdotes, Phaskell offers insights into programming languages, development frameworks, and best practices in software engineering. Readers can learn about software design principles, development methodologies, and emerging technologies to enhance their coding skills and build high-quality software applications.

Planet Haskell

A semantic project navigator tool uses topic modeling and spectral clustering to organize repository files by meaning rather than directory structure. The tool embeds files as vectors, recursively clusters them, and generates distinctive labels using LLMs. Key implementation details include tuning-free spectral clustering (capping at 20 sub-clusters), labeling sibling clusters together for distinctiveness, constraining label sizes (2 words for clusters, 3-7 for files), and incorporating path patterns. The approach works on any text documents and can be extended to images and PDFs using multimodal models.

Haskell for all: Browse code by meaning