A Study on Clustering of SNS SPAM using Heuristic Method
Journal: The Journal of the Institute of Internet, Broadcasting and Communication (Vol.14, No. 6)Publication Date: 2014-12-31
Authors : Young-Man Kwon; In-Rak Lee; Myung-Gwan Kim;
Page : 7-12
Keywords : SNS; Python; Twitter; Spam; Spammers; Classification;
Abstract
It has good features for social networking with friends SNS is maintained. However, various enterprises, individuals invading the inconvenience spammers have exposure to a number of users to tweet spam. The study was conducted in the existing research on these spam tweets. However, the results showed a more accurate classification and detection is difficult because of the lack of precision and different causes. In this paper, we describe how to classify the characteristics of spammers, classification criteria. Also has a link rate and difference between followers and following, these features were present classification criteria for spammers account. This experiment was performed according to the criteria. Randomized trial of spam and non-spam accounts were selected and account type was conducted according to the criteria 68% of the link ratio of spam accounts. Followers / Following ratio was 27581.5. Non-spam accounts was 6.12%. Followers / Following ratio was 1.26.
Other Latest Articles
- Design of Smart Service based on Reverse-proxy for the Internet of Things
- Applied Method of Energy Harvesting for Multi-Relay Environment
- Network-Based Partially-Distributed Mobility Management Mechanism and Performance Evaluation
- Distributed Dynamic Lighting Energy Management System based on Zigbee Mesh Network
- Bicycle Accident Position Tracing and Alarm System
Last modified: 2016-01-19 11:30:09