Towards Automated Recommendations for Taxonomy Expansion

The partner organization is building a knowledge graph to integrate the knowledge in the SharePoint documents/pages of each domain, which will be used as the central information base to support various information retrieval tasks such as intelligent chatbots that can automatically identify relevant information. A critical element of the knowledge graph is a taxonomy of the entities (keywords) extracted from the SharePoint documents/pages, which can enrich the relations between the entities. However, with the explosive increase of entities, maintaining the taxonomy becomes beyond the capacity of the taxonomists. Thus, this project aims to develop a prototype tool to assist taxonomists in expanding the taxonomy with new entities. The tool accepts a new entity as its input, analyzes its relationship with the existing taxonomy, and outputs the recommendation for how to integrate the entity. The tool is expected to save the analysis time of the taxonomists by 50%.

Faculty Supervisor:

Heng Li;Foutse Khomh

Student:

Partner:

Banque Nationale du Canada

Discipline:

Computer science

Sector:

Finance and Insurance; Professional, scientific and technical services

University:

Polytechnique Montréal

Program: