Year: 2020
Type: article
Abstract: Online opinion spamming has become a potential threat widespread in this digital \nera as most decisions from the purchase of a simple product to consulting certain doctor \nare taken based on the online user opinions. Taking this as an advantage, businesses in \nvarious fields have either committing online spamming or being affected by the same for \nseveral reasons like market competition and profit gains. \n \nDespite of the significant research carried out in identifying spam reviews, there is \na huge gap left unbridged in detecting the spamming activity on the business as a whole \n(we measure this as honesty of the businesses). Identifying a single review to be spam or \nbenign cannot clearly justify the business to be dishonest or trustworthy. With the \nadvancements in the camouflage strategies followed by malicious users (spammers) in \nwriting fake reviews, it has become difficult to categorize a review as a spam/no-spam. \nOne such important strategy is singleton review technique ??? the technique where reviewers \ncreate multiple accounts and write only one review under each account. A large number of \nsuch Singleton Reviews (SRs) constitute to a biased review of the overall business. Recent \nresearch reveals that singleton reviews are a significant source of spam reviews and largely \naffects the ratings of online businesses. For example, about 68% of the amazon review data \nare singleton reviews. \n \nIn this research project, we focus on detecting the businesses that are affected by \nopinion spamming over time. We take advantage of the Yelp review data containing \nreviews from 5,044 business by 260,277 reviewers. We leverage the recent techniques in \ndeep learning such as transfer learning, semantic embeddings, auto encoding and LSTMs \nto classify the business as honest or dishonest based on semantic analysis of their reviews \nover time. \n \nExtensive experiments showed that the proposed models outperformed the baseline \nmodels in terms of precision, recall, and F1 score metrics in identifying both honest and \ndishonest businesses.
Author: Praveen Kumar Reddy Hande
Language: English
Topic: Spam and Phishing Detection
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3215199863