Consumer Rating Analysis and Recommendation Methodology Using Machine Learning

Reddy, Sainandan and Khan, Aayan and Saxena, Shikhar and et al, . (2022) Consumer Rating Analysis and Recommendation Methodology Using Machine Learning. In: 2022 International Mobile and Embedded Technology Conference, MECON 2022, 10 March 2022 through 11 March 2022, Noida.

Text
2022_International_Mobile_and_Embedded_Technology_Conference.pdf - Published Version
Restricted to Registered users only
Download (259kB) | Request a copy

Abstract

In the era of online shopping, Online platforms are loaded with online shopping. Nowadays, online shopping has become a trend and a necessity of fast delivery to home. Online platforms sites like Twitter, in this new time, is stacked with product marketing and feedback information. One of the most broadly utilized miniatures writing for a blog webpage, Twitter, is the place where individuals share their thoughts as tweets and thusly it becomes perhaps the best hotspot for nostalgic investigation of feedback of products. Assessments can be generally assembled into three classifications for positive, negative and neutral and the method involved with dissecting contrasts of conclusions and gathering them in this large number of classes is done by data modeling. Data mining is fundamentally utilized to uncover applicable data from tweets information particularly from the informal communication destinations. A Model is also designed with Micro blogging solution for the evaluation of reviews. In many social sites there are many online marketing recommendation may be done with recommendation post from the product consumer which would initiate the growth of any business and also eliminate the need of any raw data for analysis of the consumer rating. In the model, we propose a methodology two methodology, one is analysis from social sites like Facebook and other social platform and another is recommendation method which decrease the marketing strategy required for the consumer to consumer connection. A testing recommendation helps to remove not required information from the product consumer to product consumer communication and this helps for the product recommendation analysis to be done properly. © 2022 IEEE.

[error in script]

IITH Creators: