Categorize balanced dataset for troll detection


EOI: 10.11242/viva-tech.01.04.160

Download Full Text here



Citation

Mr. Prashant Singh, Ms. Nidhi Singh, Mr. Namit Rasalkar, Prof. Pallavi Raut, " Categorize balanced dataset for troll detection", VIVA-IJRI Volume 1, Issue 4, Article 160, pp. 1-6, 2021. Published by Computer Engineering Department, VIVA Institute of Technology, Virar, India.

Abstract

As we know Cyber bullying is increasing day by day and Cyber troll is one of the cyber-aggressive actions that is not much different from cyberbullying in online abuse so that the victims feel uncomfortable. One of the most used social media platforms in which cyber trolling frequently happens is Twitter. Basically, it is found that during an investigation of cyberbullying cases a lot of information gathered is false which aims to give discomfort, hatred and waste lots of time. So, it is necessary to classify between cyberbullying tweets and normal tweets on twitter. There has already been research on classification of cyberbullying tweets and normal tweets using the Support vector machine (SVM) algorithm. But the drawback of the system is that it only gives 63.83% of accuracy. Firstly, we can improve the accuracy of the system by using the Recurrent Neural Network (RNN) And Secondly, for balancing the dataset we will be using Synthetic Minority Over-sampling Technique (SMOTE). We believe that using these techniques we will be able to increase the accuracy of the previous proposed.

Keywords

Cyber bullying, Twitter, RNN, SVM, SMOTE.

References

  1. Ahmed AL Marouf, Rasif Ajwad, Adnan Ferdous Ashraf, “Looking Behind the Mask: A framework for Detecting Character Assassination via Troll Comments on Social media using Psycholinguistic Tools.”, International Conference on Electrical, Computer and Communication Engineering (ECCE), 2019.
  2. Wanda Athira Luqyana, Beryl Labique Ahmadie, Ahmad Afif Supianto, “K-Nearest Neighbors Undersampling as Balancing Data for Cyber Troll Detection”, IEEE, 2019.
  3. Patxi Galan Garcia, José Gaviria de la Puerta, Pablo Garcia Bringas, “Supervised machine learning for the detection of troll profiles in twitter social network: application to a real case of cyberbullying”, http://jigpal.oxfordjournals.org/, last accessed on: 3/11/2020.
  4. Ushma Bhatt, Divya iyyani, Keshin Jani, Ms. Swati Mali, “Troll-detection systems Limitations of troll detection systems and AI/ML anti-trolling solution”, International Conference for Convergence in Technology - IEEE, 2018.
  5. Todor Mihaylov, Preslav Nakov, “Hunting for Troll Comments in News Community Forums”, 54th Annual Meeting of the Association for Computational Linguistics, 2016.
  6. Amit Pratap Singh, Maitreyee Dutta, “Spam Detection in Social Networking Sites using Artificial Intelligence Technique”, International Journal of Innovative Technology and Exploring Engineering (IJITEE), 2019.
  7. Dr. Priyanka Harjule, Astha Gurjar, Harshita Seth, Priya Thakur, “Text Classification on Twitter Data”, IEEE, 2020.