ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Performance Improvement in Multimedia Answering By Web Excavation?

Journal: International Journal of Computer Science and Mobile Computing - IJCSMC (Vol.3, No. 5)

Publication Date:

Authors : ; ;

Page : 1191-1197

Keywords : Multimedia QA; Question classification; Machine Learning;

Source : Downloadexternal Find it from : Google Scholarexternal


Question Answering (QA) can be considered as an alternative of Information retrieval systems. It is a way of responding to a query which is asked in natural language, with accurate and precise result. The relevance of question answering lies in the downside of search engines, which return a lot of irrelevant documents based on some key terms. Community based Question Answering (CQA) services are defined as dedicated platforms for users with diverse background to share information and knowledge and to respond to other users’ questions, resulting in the building of a community where users interactively give ratings to questions and answers. But the downside of existing CQA forums are, most of the previous systems are text based and fail to provide more detailed information which helps the user to understand the things completely. Here I propose a model that is able to provide answers from different CQA forums along with suitable multimedia information. In this model, first the combination of media through which question should be answered is selected based on question and answer pairs, in the next stage the most relevant keyword will be selected based on question & answer, then in the final stage it will collect appropriate multimedia information from different web sources and presented to the user along with textual information. Compared to lot of multimedia question answering approaches, it mainly focus on extracting textual answering from different web sources along with the multimedia information and it is faster compared to other approaches, and the result should be precise and appropriate media data.

Last modified: 2014-05-31 19:37:22