QUORA INSINCERE QUESTIONS CLASSIFICATION

Abstract

The competence of any online website surely depends on the kind of experience it gives to its users which depends by and large on the content they put up on their website. Hence, the content which is being put online should be really taken care of. There are many websites which provide content to their user in terms of questions and answers for example the online website Quora, which has large scale of data in terms of questions and answers of users. Users are the ones who put up questions and also provide answers to those questions. Most people on Quora are well-meaning and are genuinely interested in asking questions. Less often, in a way that is deliberately provocative, somebody will ask a question where the wording is intended to make its own statement. This may include framing or calls for hateful stereotypes to be confirmed. These questions are harmful to our community, and we remove or hide them whenever we become aware of them. In this paper a system is proposed that will take significant amount of data from quora and use that data for different approaches to predict if the question is insincere. This project aims to develop models that take the text of a question as an input in English and produce a o or 1 that corresponds to whether the question should be approved as “sincere” or flagged as “insincere”.