Open AccessToxic Comment Classification
Author(s) -
Sonika Prakash
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.36860
Subject(s) - computer science , constructive , conversation , convolutional neural network , social media , limit (mathematics) , artificial intelligence , transformation (genetics) , noise (video) , identity (music) , natural language processing , machine learning , world wide web , sociology , communication , image (mathematics) , mathematics , mathematical analysis , biochemistry , chemistry , physics , process (computing) , acoustics , gene , operating system
A large proportion of online comments available on public domains are usually constructive. However, a significant proportion is toxic and destructive. Several platforms and social media sites are finding it difficult to maintain fair conversation and are often forced to either limit the user comments or get dissolved by shutting down user comments completely. So, to prevent these types of identity hate through comments on social media, we come up with a solution to detect different types of toxicity in the comments using Deep Learning and Natural Language Processing. Dataset is obtained online which is processed to remove noise. Transformation of raw comments is done before feeding it to the classification model using Natural Language Processing. A Convolutional Neural Network model is used which will differentiate toxic comments from non-toxic comments.