An optimal support vector machine based classification model for sentimental analysis of online product reviews
| Autor: | R. Ponnusamy, M. Aramudhan, P. Vijayaragavan |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Computational model
Computer Networks and Communications End user Computer science business.industry Sentiment analysis Confusion matrix 020206 networking & telecommunications 02 engineering and technology Machine learning computer.software_genre Fuzzy logic Purchasing Support vector machine Identification (information) Hardware and Architecture 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Product (category theory) Artificial intelligence business Cluster analysis computer Software Soft set |
| Zdroj: | Future Generation Computer Systems. 111:234-240 |
| ISSN: | 0167-739X |
| Popis: | In present days, recent developments in data analytics take place that allows the identification of underlining trends through effective computational models. In several e-commerce and social platform, massive number of online product reviews is posted by end users that significantly help the developers with priceless insight while designing the products. This paper presents a new cluster based classification model for online product reviews. The presented model comprises several processes. Initially, support vector machine (SVM) based classification model is applied to classify the product reviews. Then, confusion matrix is generated to consider the possibilities of every consumer purchasing the product. Next, K-means clustering technique is applied to cluster the available data into two groups. At the next stage, sentimental analysis approach is employed to extracting the features. Finally, fuzzy based soft set theory is applied to determine the possibility of the customer to purchase the product effectively. The experimental validation of the presented model takes place on ipod dataset. The simulation outcome pointed out the superior characteristics of the presented model under several aspects. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect