2020-10-09

2020-10-09

Meeting with: Abhigyan, Zubair

Basics of what we need:

• preprocessing line:
  • (OUTPUT FORMAT I1)
  • Looking at the eval data and figuring how to preprocess that.
• Setup the QA generation tool:
  • Look at how it inputs data (INPUT FORMAT I1)
  • For that reason this needs to be done first
• Text-summariser:
  • INPUT FORMAT I1: Converter from I1 to whatever the summariser needs
  • OUTPUT FORMAT I1
• BERT-Extractive summariser:
  • INPUT FORMAT I1: Converter from I1 to whatever the summariser needs
  • OUTPUT FORMAT I1

Own summariser:

• Collect Data:
  • NCERT:
    • Informative: Bio texts
    • Narrative: English Literature
• SVM
• Decision Trees
• Spacy

Splitting the tasks in order:

• QA tool
• preprocessing
• two summarizers
• data collection