Cyber Bullying Detection Using Machine Learning

Abstract:

Cyber bullying is the process of sending wrong messages to a person or community which causes heated debate with users. Cyberbullying is mostly seen in social networking sites where users reply to post with bullying words to threaten or insult other users. Cyberbullying is considered a misuse of technology. According to the latest survey done on all over the world data day by day, cases are increasing on cyberbullying. In order to solve this problem many natural language processing techniques are proposed by various authors which are time taking and not automatic. With the advancement of machine learning and artificial intelligence, models can be created and automatic detection can be implemented. To show this scenario live chat application is developed in python programming with multiple clients and one server and the Naive Bayes algorithm is used to train the model on a Twitter dataset and using this model live detection of cyberbullying is predicted and alert messages are shown on the chat application.

Problem statement:

       Social networking and online chatting application provide a platform for any user to share knowledge and talent but few users take this platform to threaten users with cyberbullying attacks which cause issues in using these platforms.

Objective:

    To provide a better platform for users to share knowledge on social networking sites there is a need for an effective detection system that can automate the process of cyberbullying detections and take decisions.

Existing system:

  • Techniques like unsupervised labeling methods which use N-gram, TF-IDF methods to detect cyberbullying are used which use the youtube dataset to detect attacks.
  • A support vector classifier is used to train models for detection.

Disadvantages:

         Techniques which are used in the existing system are not automated they need time to process request and update response.

           Social networking and chatting sites require automated detecting and processing methods.

Proposed system:

         Cyberbullying detection is designed using machine learning techniques. Twitter data set is collected with features and labels and mode is trained using the Naive Bayes algorithm and trained model is applied to live chatting application which has multiple clients and a single server. For each message, cyberbullying is detecting using the model and then alert messages are posted on chat boards.

Advantages:

         Cyberbullying detection process is automatic and time taken for detection is less and it works on the live environment. 

            The latest machine learning models are used for training models that are accurate.

Software Requirement:

Programming language: python

Front End GUI : tkinter

Dataset: Twitter cyberbullying dataset

Algorithm                       : Naive bayes

Leave a Reply

Your email address will not be published. Required fields are marked *