A Simple Framework to Mine Insights from Natural Language Model

We want to support the process of interpreting learned characteristics of Natural Language Embeddings. We aim to shed light on covered and missed semantics using a knowledge base.

Another key contribution is it to map dimensions to semantics or facts.

Contents:

• Evaluation Pipeline
• Metrics
• Tasks Collection
  • Submodules
  • tasks.analogy_task module
  • tasks.neighborhood_baseline module
  • tasks.neighborhood_task module
  • tasks.similarity_task module
  • tasks.task module
  • Module contents
• Data Source Module
  • sources.CSVSource module
  • sources.source module
  • Module contents
• Visualization Tools
  • Submodules
  • visualization.similarity_visualizer module
  • visualization.visualizer module
  • Module contents

Indices and tables

• Index

• Module Index

• Search Page