An Ensemble Approach for Cyber Bullying: Text Messages and Images

Abstract

Text mining (TM) is a domain used to find valuable patterns from various text documents. Cyberbullying is the term used to abuse a person online or offline platform. Nowadays, cyberbullying has become more dangerous to people who are using social networking sites (SNS). Cyberbullying is of many types, such as text messaging, morphed images, videos, Etc. It is a challenging task to prevent this type of abuse of the person in online SNS. Finding accurate text mining patterns gives better results in detecting cyberbullying on any platform. Cyberbullying is developed with the online SNS to send defamatory statements or orally bully other persons, or by using the online forum to abuse in front of SNS users. Deep Learning (DL) is one of the significant domains used to extract and learn the quality features dynamically from the low-level text inclusions. In this scenario, Convolution neural network (CNN) are DL models used to train text data, images, and videos. CNN is a compelling approach to preparing these data types and achieving better text classification. This paper describes the Ensemble model with the integration of Term Frequency (TF)Inverse document frequency (IDF) and Deep Neural Network (DNN) with advanced feature-extracting techniques to classify the bullying text, images, and videos. Feature extraction technique extracts the features of cyber-bullying patterns from the text and images. A limited number of datasets are used to classify the data. The proposed approach also focused on reducing the training time and memory usage, which helps the classification improvement.