Weekly Update 1: Co-reference resolution and word2vec approach to improve extractive summarization

Progress

  • implementing "summarising multiple notes simultaneously" feature
  • users can now summarize multiple highlighted texts in notes
  • understanding co-reference resolution:
  • understanding word2vec solution to create similarity matrix:
    • semantic relationships between sentences
  • discovered Javascript NLP library: Natural
    • docs: Introduction | Natural
    • useful for tokenizing text into sentences and has default PageRank algorithm (useful for custom implementation of TextRank)

Plans

  • implementing "summarising notebooks" feature
  • learn more about WASM
  • making a table of all extractive summarisations and discover their strengths and weaknesses (still pending)
  • if all the above are done: try to implement and test Hobb's Algorithm (non-ML co-reference resolution)

Problems

  • I am still having issues with setting up Transformers.js in a plugin.
1 Like