The Consequence of Fake Analyses on E-Commerce During and After Covid-19 Epidemic: SKL-Based Fake Analyses and Recognition

The Consequence of Fake Analyses on E-Commerce During and After Covid-19 Epidemic: SKL-Based Fake Analyses and Recognition

RAJALAKSHMI.C[1],

ASSISTANT PROFESSOR,

R. B. GOTHI JAIN COLLEGE FOR WOMEN

USHA J[2]

ASSISTANT PROFESSOR,

R. B. GOTHI JAIN COLLEGE FOR WOMEN

Abstract:

The epidemic of Covid-19 and the prosecution of lockdown, public isolation, and other defensive procedures lead to a global intensification in online shopping. The growing consequence of online shopping and widespread use of e-commerce has improved race among companies for online marketing. Highpoints that online evaluations show a important role in enhancing a business or smearing it. Product review is a vital feature in customers’ decision-making, foremost to an powerful subject known as duplicitous or fake reviews recognition. Given these analyses’ influence over a business, the traitorous acts of giving false analyses for individual improvements have enlarged with time. In our research, we proposed a fake review recognition model by using Text Sorting and techniques related to Machine Learning. We used classifiers such as Support K-Nearest Neighbor, logistic regression (SKL), and Vector Machine using a bigram model that detects duplicitous reviews based on the number of pronouns, sentiments, and verbs. Our future methodology for detecting fake online reviews outperforms on the squeal dataset and the TripAdvisor dataset compared to other state-of-the-art techniques with 95% and 89.03% accuracy.