Automatic Tagging for Document Retrieval in Corporate Environments

Knowledge-intensive industries accumulate content at a rate and volume that is impossible for humans to manually organize and retrieve. A typical search engine based on keywords is not effective, because the user must provide the exact words contained in the text of the document or description of an entity and cannot filter search results based on the implicit context of the search. The search capabilities of the search engine can be extended by content metadata in the form of tags. However, manually adding metadata to each content element has prohibitive cost. Oris4 is a product that aims to provide easy access to corporate content by automatically extracting the required metadata for search. The proposed project aims to apply recent research results in automatic tagging of documents in large corpora towards substantially enhancing the quality of metadata in Oris4 and thereby making it a more competitive product in the marketplace.

Faculty Supervisor:

Vlado Keselj

Student:

Partner:

2nd Act Innovations Inc

Discipline:

Computer science

Sector:

Information and cultural industries

University:

Dalhousie University

Program: