Skip to content
This repository was archived by the owner on Nov 1, 2025. It is now read-only.

Conversation

@sadmanca
Copy link
Member

No description provided.

sadmanca added 16 commits June 22, 2025 22:33
…tion extraction)

Introduces several advanced NLP methods for creating more semantically rich knowledge graphs from research papers.

- Adds new node creation strategies:
  - Scientific entity extraction using a transformer-based NER model (SciBERT).
  - High-level concept extraction using a hybrid of regex, NLTK, and TF-IDF to identify methods, models, and challenges.

- Implements new edge creation techniques:
  - Relation extraction using spaCy to identify subject-verb-object triples between concepts.
  - A sophisticated method that combines semantic similarity, co-occurrence analysis, and contextual pattern matching to create typed relationships (e.g., "uses", "improves").

- Provides a new unified `generate_knowledge_graph` API to easily select and combine different extraction methods.

- Enhances graph visualization to display relationship labels on edges, a legend, and network statistics.
…nforcement learning graph without unrelated papers
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants