In this day and age, surveys on web-based sites
play an important role in product sales because people try
to get all of the advantages and disadvantages of any item
before purchasing it because there are various options for a
similar item, such as different makes for a similar type of
item, or differences in merchants that can provide the item,
or differences in the method used to purchase the item, so
the audits are important, because it's difficult for them to
personally verify each item and sale, a tool called Fake
Review Detection is used to detect any fraud. The client
made the request only based on the rating and examining
the audits associated with the specific item. Others'
comments provide a wellspring of satisfaction for the new
goods customer. It's possible that a single unfavourable
audit will persuade a customer not to buy that item. In the
current situation, it's possible that this one audit is bogus.
Thus, to eliminate phoney audits and provide clients with
the first surveys and ratings associated with the items, we
proposed the Fake Product Review Monitoring and
Removal System (FaRMS), which is an Intelligent Interface
that takes the Uniform Resource Locator (URL) associated
with Amazon, Flipkart, and Mynntra results and dissects
the surveys, providing the client with the first appraising.
The suggested framework is unique in that it works with
three web-based company websites rather than only
breaking down surveys in English. The requested project
was completed successfully. The accuracy of 87 percent in
recognizing counterfeit surveys written in English was
achieved using acute learning methods, which is higher
than the precision of previous models.
Keywords : Fake Reviews Detection,, Machine Learning