Optimal neural network to enhance classification accuracy for mining online reviews and opinions using improved PSO
by B. Dhanalakshmi; Arumugam Chandra Sekar
International Journal of Networking and Virtual Organisations (IJNVO), Vol. 18, No. 4, 2018

Abstract: While using rapid growth from World Wide Web, there is volatile improve in the user-produced subject matter such as purchaser evaluations, websites, discussion community forums, social networks and so forth In previous work, we have implemented opinion mining using three phase. They are: 1) data pre-processing; 2) opinion extraction; 3) opinion mining. In feature extraction, features, such as term frequency, part of speech (POS), syntax, negation and term-based features beyond term unigrams were extracted from the words in documents. In the final step, ranking and classification was done. In the present work, we will implement the same three phases as previous work but with different process in each of the following steps such as: 1) data reprocessing; 2) opinion extraction; 3) opinion mining. The second phase will be classified into two, i.e., feature extraction and opinion extraction. After feature extraction, we extract useful information related to item's features and use to rate them as positive, neutral, or negative.

Online publication date: Tue, 31-Jul-2018

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Networking and Virtual Organisations (IJNVO):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com