Capstone Project – II Toxic Words Classification

Details about the files

Contents

Technical Summary: capstone_technical_summary.pdf

code: cap_phase_1.ipynb

Problem Statement:

With the increasing use of online platforms for communication, the prevalence of toxic language and online abuse has become a major concern. To promote a safe and inclusive online environment, there is a need for an efficient and effective toxic words classification model that can identify and categorize toxic language in comments. The aim of this capstone project is to develop a model that can accurately classify comments into toxic and non-toxic categories, which can be used to improve content moderation, sentiment analysis, and chatbot performance.

Data DEscription

Data Preprocessing

Traditional models

Tune Logistic regression

Neural network,CNN,and LSTM

Tune CNN

Pretrained Models:GloVe

Interpret the GloVe model

Conclusion