Home
Call For Papers
Submission
Author
Registration
Publications
About
Contact Us

  Detection of Spammer Group Using Semi-Supervised Learning  
  Authors : Bindhya Babu; Sam G Benjamin
  Cite as:

 

These days, online items survey assumes a pivotal job for purchasing online products. A high extent of positive surveys will bring considerable deals development while negative surveys will cause deals misfortune. Driven by huge money related benefits, various spammers attempt to advance their items or downgrade their rivals' items by posting fake and one-sided online surveys. Existing works extract spammer candidates and remove spammers from the review data using unsupervised spamicity positioning techniques. All things considered, as indicated by past research, marking few spammer group is simpler than one expect, number of techniques endeavor to utilize significant named information. In this paper, we propose a semi-supervised learning technique to distinguish spammers. Naive Bayesian model and EM calculations are used to organize a classifier for the detection of spammer groups.

 

Published In : IJCSN Journal Volume 8, Issue 3

Date of Publication : June 2019

Pages : 213-216

Figures :01

Tables : 01

 

Bindhya Babu : received her B. Tech (CSE) degree from University of Kerala in 2017. She is currently pursuing her Masters in Computer Science & Engineering from KTU.

Sam G Benjamin : is working as Assistant Professor in Computer Science and Engineering Department. His research interests focuses on image processing, data mining and image mining. He has published papers on image processing.

 

Naive Bayesian Model, EM Calculations, Spammer Groups, Semi-Supervised Learning

As individuals and businesses are progressively utilizing surveys for their choices making, it is critical to distinguish spammers who write counterfeit surveys. This paper proposed a successful strategy to distinguish spammer groups who cooperate to write counterfeit surveys. First, the PSGD model uses frequent item mining (FIM) to discover spammer group candidates from the review data. Then, by manually labeling some spammer groups as positive instances, the PSGD employs PU-Learning to construct a classifier from the positive and unlabeled instances to identify the real spammer groups from group candidates.

 

[1] Detecting Spammer Groups from Product Reviews: A Partially Supervised Learning Model by LU ZHANG, ZHIANG WU, (MEMBER, IEEE), AND JIE CAO, 10.1109/ACCESS.2017.2784370, IEEE Access [2] Detecting Group Review Spam by Arjun Mukherjee, Bing Liu, Junhui Wang, Natalie Glance, Nitin Jindal, Dept of CS. Technical Report, UIC, 2011. [3] Detection of Fake Review and Brand Spam Using Data Mining Technique by Miss.Rashmi Gomatesh Adike, Prof. Vivekanand Reddy, Volume 02, Issue 07; July - 2016 [ISSN: 2455-1457] [4] H. Li, B. Liu, A. Mukherjee, and Natalie Glance, "Spotting fake reviewer groups in consumer reviews," Computación y Sistemas, vol. 18, no. 3, pp. 467-475, 2014. [5] L. Akoglu, R. Chandy, and C. Faloutsos, "Opinion fraud detection in online reviews by network effects." ICWSM, vol. 13, pp. 2-11, 2013.