Adaptive Classification for Spam Detection on Twitter with Specific Data
Authors:
- 1 November 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Abstract
At present, the popularity of Twitter is more and more increasing. Many users can find information that tweets to Twitter. Some information is beneficial, and at the meantime, some information is created from spammers who would like to promote their websites or services. They are harmful to normal users by using the Twitter channel to exploit common user interests, such as sharing malicious links and ads. In order to stop spammers, many researchers have proposed tools to filter those junk. However, the focus of recent works is how to create streaming spam detection methods. In this paper, we proposed the adaptive data classification for spam detection by using spam word lists and a commercial URL-based security tool. We analyzed data by Naïve Bayes algorithm with both data types including all data and specific data. It can help to improve the performance of the spam detector is better than usual. We can show our proposed methods fulfillment in the experiments result.Keywords
This publication has 7 references indexed in Scilit:
- Towards Extracting Drug-Effect Relation from Twitter: A Supervised Learning Approach2016
- A survey on behaviors exhibited by spammers in popular social media networks2016
- Efficient spam detection across Online Social Networks2016
- A Performance Evaluation of Machine Learning-Based Streaming Spam Tweets DetectionIEEE Transactions on Computational Social Systems, 2015
- 6 million spam tweets: A large ground truth for timely Twitter spam detection2015
- An integrated approach to spam classification on Twitter using URL analysis, natural language processing and machine learning techniques2014
- A Local-Concentration-Based Feature Extraction Approach for Spam FilteringIEEE Transactions on Information Forensics and Security, 2010