Quora Insincere Questions Classification
| Author(s) | : | Daksh Suraj Rawat, Akshay Mungekkar, Prateek Nima |
| Institution | : | Electronics and Telecommunication, UCOE Mumbai |
| Published In | : | Vol. 8, Issue 1 — January 2021 |
| Page No. | : | 20-27 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
Internet has opened up a new horizon of opportunities. It has answers to all questions, and even if answer isnot found there are sites where you can ask any question and people from all over the world answer it. Quora is one suchquestion answer website. Here anyone can ask question from any domain, which are then answered by others. Peoplemisuse this boon and create a havoc by asking questions that are not to be asked in a public forum. In this project wepropose to develop a system that recognizes these “insincere” questions so that appropriate actions can be taken againstthem. The questions are classified as insincere if they have a non-neutral tone, is disparaging, inflammatory or thequestions are in negative sexual aspects and if it is not grounded to reality. In this project a model of text classificationwill be developed using Natural Language Processing. Based on the models, Decision Tree performed the best followedby Random forest. The metrics used for judging the result include F1 score, recall, precision and Area under Curve.
Daksh Suraj Rawat, Akshay Mungekkar, Prateek Nima, “Quora Insincere Questions Classification”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 8, Issue 1, pp. 20-27, January 2021.
