Identification of Inappropriate Comments in Social Media

Identification of Inappropriate Comments in Social Media.

Topic of the project Identification of inappropriate comments in social media through sentiment analysis. Objective of the project Perform Sentiment Analysis on social media comments. Identify targeted hate comments or inappropriate comments. Segregate them for manual checking..

Motivation of the project Social media has become widely used around the world. With this the presence of hate spreading users has also increased. Thus several cases of cyberbullying is seen. This motivated me to make this project which protect social media users from such hate or inappropriate behaviour in online platforms. Application area In social media platforms. In online blogging pages. In product or service review systems..

Review on papers.. SN Topic of the Paper and year Work done Result obtained 1. The Decision Tree Algorithm on Sentiment Analysis: Russia and Ukraine War. 2023 Find sentiment of twitter comments in Russia Ukraine war using decision tree algo. 85.61% accuracy for data with 80-20 partition. 2. Implementation of K-Nearest Neighbor (K-NN) Algorithm for public sentiment analysis of Online learning. 2021 Public sentiment analysis of Online learning using K-Nearest Neighbor algo. 84.93% accuracy with k value equal to 10.

Review on papers.. SN Topic of the Paper and year Work done Result obtained 3. Sentiment Analysis of Madura Tourism Opinion Using Support Vector Machine(SVM). 2023 Sentiment Analysis of tourism Opinion using SVM. 92.592 % accuracy with C=1 and gamma=0.5. 4. Sentiment Analysis using Recurrent Neural Network. 2018 Performed SA accuracy using RNN and Word2Vec 91.98% accuracy when both RNN and Word2Vec was used.

Merits of the work. With this work in place social media can be free of hate comments. The incidence of cyberbullying will decline, which will lessen the impact of cyberbullying on teenage depression. The project uses deep learning and machine learning methods which will give us a better understanding of human sentiments through their comments..