Automatic Text Summarization for Code-Mixed Language using Fuzzy Logic

Natural Language Processing (NLP) is a domain of computing and natural linguistics. It is related to the unique conversation among machines and languages used by human. We are leaving in the era of machines, data and intelligence where the amount of information on the Internet is raising continuously, the method which are competent to build it preserved for users, e.g., by text summarization, question answering chat bots etc. The difficulty is highlighted with news stories, where numerous news providers account on same measures by same facts. Application like Automatic text summarization can resolve this challenge. By creating a wide-ranging summary of a given set of documents, Summarization can deal with the problem of information load through retrieving and showing the main content. This is still increased via the escalating quantity of information in community media where collection opinion in diverse languages can be verified. This paper addresses the boundaries of summarization by means of upward techniques that work in an extremely tall quantity and for multilingual scenario. Still the evaluation of automatically formed summaries in diverse languages is a exigent problem for the summarization community, as because human hard work are multiplied to make model summaries for each language. The retention ratio for above method is found to be around 75%.