Query Modulation For Web-Based Question Answering

Abstract

Abstract:The web is now becoming one of the largest information and knowledge repositories. Many large scale search engines (Google, Fast, Northern Light, etc.) have emerged to help users find information. In this paper, we study how we can effectively use these existing search engines to mine the Web and discover the -correct'' answers to factual natural language questions. We propose a probabilistic algorithm called QASM (Question Answering using Statistical Models) that learns the best query paraphrase of a natural language question. We validate our approach for both local and web search engines using questions from the TREC evaluation.

Domain and Intelligence Based Multimedia Question Answering System

“…However, the resemblance estimation actually should be question adaptive. For example, if we want to find a person, we should calculate the similarities of facial skin texture instead of the features extracted from the whole images [17]. It is sensible as information seekers are future to find a person rather than other objects.…”

Multimedia Based Community Question Answer by Harvesting Web Based Information

“…In question based classification, need to extract core sentence from each question [16] .This classification contains two steps. First question can be categorized according to interrogatives word and some question directly gives the answer with text.…”

“…If given query is related to person, then face detection method is used to identify images return by search engine. If person related query are found in images or key frame, which is extracted by 256-D local binary pattern feature from largest faces of image or video frames [16]. Non person related queries extract 428 dimensional global visual features after that reranked performed on removal of duplicate image.…”

Language Independent Answer Prediction from the Web

“…However, scaling this new QA technology to the Web in order to improve current search engines to efficiently locating information presents extraordinary challenges, cf. [2]. Consider for example the enormous size of the Web content that is currently indexed by the best search engines (Billions of Web pages).…”