期刊名称：International Journal of Advances in Engineering and Management
电子版ISSN：2395-5252
出版年度：2022
卷号：4
期号：5
页码：720-725
DOI：10.35629/5252-0405424427
语种：English
出版社：IJAEM JOURNAL
摘要：Machine Learning is one of the fastest growing research study area, which allows customer to make better-informed buy selections appropriate recognizing and analysis of reviews from the internet and social media. In lots of online sites, there are choices for publishing reviews, and therefore producing scopes for fake paid reviews or untruthful reviews. As purchasers can't ask about an item or assess previously purchasing from on-line, they check out reviews and after that choose to purchase some products. Although they looking for reviews throughout various sites, they might not have the ability to determine whether it is a spam review or not. Some business with their Social Media Optimization group include some great reviews by themselves in purchase to make the item popular. So the business will provide fake great review for various items produced by their very own business. So the individual will unable to discover whether the review is true or not . Recently, the spam review detection issue has acquired a lot interest from communities and scientists, however still they have to carry out experiments on real-world massive review datasets. This can help to reviews the effect of extensive opinion spam in on-line reviews. Bulk of present research study has focused on machine learning techniques, which need identified information - an insufficiency when it concerns online reviews. Amazon.com on-line reviews text information is utilized for categorizing the item reviews. Lots of traditional machine learning algorithms have been executed to categorize the reviews .However still there's range to enhance the precision. In order to achieve betterprecision Nave Bayes formula is to be executed to categorize the reviews as spam and authentic reviews.
关键词：Fake reviews;NLP(Natural Language Processing);Machine Learning;Machine Learning;Word Embedding;CountVectorizer
