|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
Sentimental Analysis has been popular in recent years with all online product companies. The number of people who utilize a specific product has grown, prompting the industry to improve the product's features. Many people who utilize websites, blogs, and online shopping these days leave reviews on the things they have used. Other people took these reviews into account while looking for items. As a result, most of the companies decided that delivering the good product to the consumer depends on user evaluations employing the emotional analysis approach. In Sentimental Analysis a group of users providing reviews are collected and processed to suggest the users. The reviews that are offered are lengthy and contain many paragraphs of substance. The data for this paper was gathered from the Amazon website. The dataset if pre-processed and categorized using the Naive Bayes and SVM algorithms. These previous methods delivered precision that was insufficient. As a result, an ensemble technique was used to improve the accuracy of the reviews. Theproposed classification method combines SVM and Naïve Bayes algorithms and calculates the mode value for each algorithm using the vote reference. We developed an Ensemble technique that improves on the current algorithm's accuracy. After the accuracy is computed, the user is suggested a certain product based on the reviews.
Keyword
Machine Learning, SVM, Naive Bayes, Ensemble, Product Reviews
PDF Download (click here)
|
|
|
|
|