Supervisor: Prof. Benjamin C.M.Kao.
Student: Ma Zhiyu; Xu Bing
In the Age of Big Data, extraction of accurate and useful information from numerous sources becomes increasingly important. Mainstream information retrieval techniques are popular search engines like Google and Yahoo. However, they are facing their limitations: search engines are unable to understand natural language question, which is also our motivative for this project. Our project is called Knowledge-based Question Answering System, which aims to implement a system that based on a knowledge database can react to some questions in natural language and provide reasonable answers.
The basic goal for our project to set up a KBQA system which can react to natural language questions. The advanced goal for our project is to further improve the feature extraction and ranker part in order for the system to offer more accurate and custom feedback.
As for our planned goal, after data training, the system should be able to answer simple questions like "What is 3 plus 5?" clearly and correctly. Furthermore, it should provide corresponding reasonable answers to ambiguous question like "What is 3 and 5" (since "and" can mean addition or multiplication) after certain trainings.
Month | Task List |
---|---|
Sept 2015 |
|
Oct 2015 |
|
Nov - Dec 2015 |
|
Jan 2016 |
|
Feb - Mar 2016 |
|
April 2016 |
|