Information extraction from real-world business documents

This research project aims at creating a robust, efficient and reliable tool for Information Extraction (IE) from vast amounts of textual data related to the financial domain. Named entities recognition, a subtask of information extraction, seeks to locate and classify elements in text into pre-defined categories such as the names of persons, organizations, locations, expressions of times, quantities, monetary values, percentages, etc. The targeted IE system will extract structured data or knowledge from the un-structured financial texts by identifying references to the named entities as well as stated relationships between such entities through coreference resolution and relationship extraction.

Faculty Supervisor:

Fatiha Sadat

Student:

Tan Ngoc Le

Partner:

Metix

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Université du Québec à Montréal