Abstract—As various review sites grow in popularity and begin to hold more sway in consumer preferences, spam detection has become a burgeoning field of research. While there have been various attempts to resolve the issue of spam on the open web, specifically as it relates to reviews, there does not yet exist an adaptive and robust framework out there today. We attempt to address this issue in a domain-specific manner, choosing to apply it to Yelp.com first. We believe that while certain processes do exist to filter out spam reviews for Yelp, we have a comprehensive framework that can be extended to other applications of spam detection as well. Furthermore, our framework exhibited a robust performance even when trained on small datasets, providing an approach for practitioners to conduct spam detection when the available data is inadequate. To the best of our knowledge, our framework uses the most number of extracted features and models in order to finely tune our results. In this paper, we will show how various sets of online review features add value to the final performance of our proposed framework, as well as how different machine learning models perform regarding detecting spam reviews.
Index Terms—Feature extraction, machine learning, predictive analytics, spam detection.
The authors are with Courant Institute of Mathematical Sciences, New York University, New York, NY 10003 USA (e-mail: firstname.lastname@example.org, email@example.com).
Cite: Junzhang Wang, Diwen Xue, and Karen Shi, "An Ensemble Framework for Spam Detection on Social Media Platforms," International Journal of Machine Learning and Computing vol. 11, no. 1, pp. 77-84, 2021.Copyright © 2021 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).