A Review

Abstract

One of the most well-liked social media is Twitter. Spam is one of the several issues that negatively affect users. The objective of this study is to provide an overview of different techniques used for detecting spam in twitter. The proposed framework mainly contains the comparison of four existing twitter spam detection techniques namely, machine learning, feature based detection, combinational algorithm, and deep learning. Machine learning detection uses techniques such as SVM, future engineering, machine learning framework, and semantic similarity function to assess spam. In feature based detection, metadata based, tweet based, user based, and graph based techniques are used to detect spammers. In combinatorial algorithm detection, Naive Bayes-SVM, K-nearest neighbour-SVM, random forest-SVM and RNN-Short term memory techniques are used to detect spam. Deep learning detection uses feature based, semantic cnn, convolution-short term memory nn, and deep learning convolution technique to identify spam. This paper covers relevant work and comparison of several anti spamming techniques.