Study on optimized information retrieval techniques for question answering system

Abstract

The internet data grows abundantly and accumulated as unstructured data such as textual documents images videos and audio content etc These information contents are generated on daily basis from various social media sources such as blogs twitter news articles community sites and Facebook etc Information sharing is happening across applications of various industries such as health care education social websites military forums community based QA sessions customer care and so on Information retrieval techniques are used for obtaining relevant information from a collection of resources to solve user needs anywhere at any time Now-a-days during web search users anticipate precise answers with short response time instead of a list of documents This necessitates in the development of new learning models enhancement algorithms to automatically retrieve answers and summarize them Data sources appear in different formats and hence it is very difficult to retrieve data without processing it for easy retrieval The structure and arrangement of data in a specified format is needed and it is achieved by data preprocessing and clustering by its similar groups More techniques can be explored in question classification answer extraction documents clustering and text summarization for the retrieval of high quality candidate answers Systematic approaches are required with the help of Natural Language Processing Machine Learning semantic similarities and artificial intelligence field etc in order to deal with large amount of data newline