Problem Statement
Due to the development of the Internet in recent years and the change in people’s habits and behavior towards technology, social networks have gained more and more popularity and users, generating an impressive amount of comments on any topic on a daily basis.Unfortunately, this also means that a significant amount of these comments may contain inappropriate content such as obscene, aggressive, rude, racist, sexist or violent phrases.
Our Purpose
In this work, we explored the use of certain natural language processing (NLP) and machine learning techniques to detect violent content in text.We have compared six models naive bayes (NB),Support Vector Machine (SVM), Logistic Regression (LR), GradientBoosting (GR), Random Forest (RF) and we have also tried a model based on deep learning called Long short-term memory (LSTM).Finally,we made a proposal on the most excellent approach for recognizing and classifying inappropriate content on text documents.
Achieved Result
we will present the human-machine interfaces of our application through several screenshots.
Authentication page
This interface describes the entry point for our web application. As shown in the figure, To sign in you must enter your Email and password.
Prediction Page
Once you login to your account you can use the prediction page,you can write a textand see the result of the classification of this text.
Result Page
