Submit your paper : editorIJETjournal@gmail.com

Paper Title : FAKE PRODUCT REVIEW SUPERVISION AND PRODUCT RECOMMENDATION

ISSN : 2395-1303
Year of Publication : 2020
10.29126/23951303/IJET-V6I2P2

Authors: -Mr. C. Mani, M C A, M.E, M Phil, Mr. N. Karthikeyan, M C A

         


Abstract
Nowadays, there are a unit variety of individual’s exploitation social media opinions to form their appeal buying product or service. Opinion Spam detection is associate exhausting and onerous downside as there are a unit several pretend or faux reviews that are created by organizations or by the individuals for varied functions. They write faux reviews to mislead readers or automatic detection system by promoting or demoting target product to market them or to degrade their reputations. Opinion spamming refers to the employment of excessive and illicit strategies, like making an outsized volume of faux reviews, so as to get biased positive or negative opinions for a target product or service with the intention of promoting or demoting it, severally. The reviews created for this purpose area unit referred to as faux, spam or fake reviews, and therefore the authors answerable for composing such deceptive content area unit referred to as faux or spam reviewers. This project can verify faux reviews created by the shoppers so block them. the subsequent things area unit thought-about within the project. 1) Tracking IP address of the user to discover if the reviews area unit from a sender. If multiple reviews area unit from an equivalent science address then the Reviews area unit thought-about Spam. 2) Exploitation Account accustomed check whether or not the reviews area unit done exploitation an equivalent account. 3) Complete solely Review detection i.e. whether or not the reviews area unit on solely complete not the merchandise. It’s not useful to think about solely the complete price to gauge a product. 4) Exploitation Negative lexicon i.e. the negative words area unit known within the review. If there is a unit quite. 5) Negative Words then the review could be a Spam.

Reference
[[1] P. Symeonidis, E. Tiakas, and Y. Manolopoulos, “Product recommendation and rating prediction based on multi-modal social networks,” in Proc. 5th ACM Conf. Recommender Syst., 2011, pp. 61–68. [2] C. Gentile, S. Li and G. Zappella, “Online clustering of bandits,” J. Mach. Learn. Res., Workshop Conf. Proc., vol. 32, pp. 757–765, 2014. [3] A recommender system for team formation in Manet. J. King Saud Univ.-Comput. Inf. Sci. 27, 147–159.Ali, F., Kwak, D., Khan, P., Ei-Sappagh, S.H.A., Islam, S.R., Park, D., Kwak, K.-S., 2017. [4] Merged ontology and svm-based information extraction and recommendation system for social robots. IEEE Access 5, 12364–12379.Bilal, M., Israr, H., Shahid, M., Khan, A., 2016. [5] Sentiment classification of roman-urdu opinions using naïve bayesian, decision tree and knn classification techniques. J.King Saud Univ.-Comput. Inf. Sci. 28, 330–344. [6] Rule growth: mining sequential rules common to several sequences by pattern-growth. In: Proceedings of the 2011 ACM symposium on applied computing. ACM, pp.956–961. [7] Fournier-Viger, P., Tseng, V.S., 2011. In: Mining top-k sequential rules In International Conference on Advanced Data Mining and Applications.Springer, pp. 180–194 [8] Jiang, M., Cui, P., Chen, X., Wang, F., Zhu, W., Yang, S., 2015. Social recommendation with cross-domain transferable knowledge. IEEE Trans. Knowl. Data Eng. 27, 3084–3097. [9] Kermany, N.R., Alizadeh, S.H., 2017. Hybrid multicriteria recommender systemusing ontology and neuro-fuzzy techniques. Electron. Commer. Res. Appl. 21, 50–64. [10] Fournier-Viger, P., Faghihi, U., Nkambou, R. andMephu Nguifo, E. CMRules: An Efficient Algorithm for Mining Sequential Rules Common to Several Sequences. In Proceedings 23th Intern. Florida Artificial Intelligence Research Society Conference (Daytona, USA, May 19-21, 2010), AAAI Press, 410-415.

Keywords
Opinion spamming, solely review detection, Tracking IP address