You are here

Machine Learning for Question Answering

Partners:

  • Gianluca Bontempi (project coordinator and Phd supervisor) Professor at MLG - Computer Science Department, ULB
  • Pierre Sener (collaborator and industrial supervisor), Mentis Consulting
  • Boris Iolis (researcher) Phd candidate at MLG - Computer Science Department, ULB

Funding: Doctiris, Innoviris, Brussels Region

Duration: 2013-2015, renewable for two additional years.

Project Overview:

Question Answering (QA) systems allow the users to ask questions in natural language, and retrieve the answers from a knowledge base, often composed of unstructured text data. Such systems can be applied in several contexts, ranging from e-Learning to commercial web pages, or simply search engine extensions. A successful QA system requires the use of advanced methods in both Machine Learning and Natural Language Processing domains, and despite the wide array of potential applications, it is still an open problem.

Project Objectives:

The objective of the project is to implement a working prototype of a QA system, improving on the existing state-of-the-art by leveraging optimized Machine Learning techniques, and to assess the system in a real business context, provided by Mentis.

A description of this project was also included in the "Horizon Recherche" newsletter (in French), available here

 

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer