Review on: Generation of Multimedia Answer by Gathering Web Information?

From past few years community Question Answering (cQA) services have gained enormous popularity. It basically allows members of community to post and answer questions and also enables general users to obtain information from a comprehensive set of properly answered questions. However, existing cQA firms usually provide only textual answers, which are not informative enough for numerous questions. Through this system, we propose an idea that is capable to enrich textual answers in cQA with appropriate and meaningful media data. This system consists of three components: answer type selection, query development for multimedia search, and multimedia information selection and presentation. This method automatically identifies which type of media information should be added with a textual answer. It then automatically collects data from the web to enrich the answer. By processing this huge set of QA pairs and adding them to a pool, this system can enable a unique multimedia question answering (MMQA) approach as users can find multimedia answers by just matching their questions with those in the pool. Distinguishing from a lot of MMQA research efforts that strive to directly answer questions with image and video data, this system is generated based on community-contributed textual answers and so it is having the ability to deal with more complex questions. Different experiments on a multi-source QA dataset have been conducted. The results actually demonstrate the effectiveness of the system’s approach.