Semantic & Behavioral Feature analysis for Detecting Fake Reviews using Machine Learning
Anjali S Dilliwala, Raghavendra G S, , ,
Affiliations 1*Student, Dept. Of CSE, Shri Dharmasthala Manjunatheshwara College of Engineering, Dharwad, India. 2: Professor, Dept. Of CSE, Shri Dharmasthala Manjunatheshwara College of Engineering, Dharwad, India.
Background: In this age of technology, online business is playing a vital role in the growth of the economy of the business. Hence, people need feedback on various products, technologies, businesses, etc. Since their opinions are the input for an individual to evaluate and adapt them. Therefore, the Review system is playing a vital role in decision making. So there arises a necessity to evaluate the reviews as well since the business units are trying to generate fake reviews to identify more clients for their products.
Methods/ Statistical Analysis: In this paper, we implement two machine learning algorithms SVM and Naïve Bayes algorithm and analyse the data and predict for the new set of data. We also compare the performance of both algorithms.
Findings: In this paper, we are trying to develop a Machine learning model which analyses the reviews on various factors and obtain the necessary features and classify the reviews as a fake or non-fake review. This helps in identifying fraudulent reviews and predicts the trustworthiness of the reviews in the future.
Improvements: The system can introduce and make available Machine learning techniques and identifying fake reviews at the earliest stage.
Anjali S Dilliwala, Raghavendra G S."Semantic & Behavioral Feature analysis for Detecting Fake Reviews using Machine Learning". International Journal of Computer Engineering In Research Trends (IJCERT), ISSN:2349-7084, Vol.7, Issue 06,pp.40-45, June - 2020, URL:https://ijcert.org/ems/ijcert_papers/V7I606.pdf,
 Esposito, Gennaro. LP-type methods for Optimal Transductive SVMs. Vol. 3. Gennaro Esposito, PhD, 2014.
 N. Jindal and B. Liu, "Review spam detection", Proceedings of the 16th international conference on World Wide Web - WWW 07 (2007), ACM, pp. 1189–1190, 2007
 Myle Ott, Yejin Choi, Claire Cardie, and Jeffrey T. Hancock. Finding deceptive opinion spam by any stretch of the imagination. Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1, HLT, 2011.
 S. Xie, G. Wang, S. Lin, and P. S. Yu, "Review spam detection via temporal pattern discovery," p. 823, 2012.
 W. Etaiwi,G.Naymat, "The impact of applying pre-processing steps on review spam detection", The 8th international conference on emerging ubiquitous system and pervasion networks, Elsevier, pp. 273-279, 2017.
 P.Rosso, D.Cabrera, M. Gomez, "Detecting positive and negative deceptive opinions using PU-learning", Elsevier, pp.1-11, 2014.
 S. Feng, L. Xing, A Gogar and Y. Choi, "Distributional footprints of deceptive product reviews". In Proceedings of the 6th International AAAI Conference on Weblogs and Social Media (ICWSM), pp. 98- 105, 2012.
 M. I. Ahsan, T. Nahian, A. A. Kafi, M. I. Hossain, and F. M. Shah, "Review spam detection using active learning," 2016 IEEE 7th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2016.
 M. Singh, L. Kumar, and S. Sinha, "Model for Detecting Fake or Spam Reviews," Advances in Intelligent Systems and Computing ICT Based Innovations, pp. 213–217, Jan. 2017.
We have kept IJCERT is a free peer-reviewed scientific journal to endorse conservation. We have not put up a paywall to readers, and we do not charge for publishing. But running a monthly journal costs is a lot. While we do have some associates, we still need support to keep the journal flourishing. If our readers help fund it, our future will be more secure.