Skip to content

Latest commit

 

History

History

kgqas

README.md

Zoning KGQAS

This is a Knowledge Graph Question Answering System it is somewhat based on the paper:

S. Aghaei, E. Raad and A. Fensel, "Question Answering Over Knowledge Graphs: A Case Study in Tourism," in IEEE Access, vol. 10, pp. 69788-69801, 2022, doi: 10.1109/ACCESS.2022.31871

Instead of tourism this is using the method for a zoning information Question Answering System (QAS).

Files

  • KGQAS.ipynb - Start here. This is a very high level view of the Zoning KGQAS working.
  • semantic_parsing.py - This ties together the entire Semantic parsing phase from the other classses the below Semantic Parsing section.
  • indexes.py - The indexes are used by the Semantic Parsing class for training and classification.
  • kg_helper.py - This is helper code for this code accessing the template generation.

Semantic Parsing

These are listed in the order of their location in the Semantic Parsing phase.

  1. Question Classification - question_classification.py
  2. Entity Linking and Class Linking - entity_class_linking.py
  3. Relation Extraction - relation_extraction.py
  4. Slot Filling and Query Execution - semantic_parsing.py

Knowledge Graph

This is located in the folder ../triplesdb.