Feature Level Mining of Online Reviews Based on a Semi-Supervised Learning Model

Minxi Wang () and Xin Li ()
Additional contact information
Minxi Wang: Sichuan University
Xin Li: Chengdu University of Technology

A chapter in LISS 2014, 2015, pp 709-715 from Springer

Abstract: Abstract Online reviews are written by customers based on personal usage experience. They not only help manufacturers better understand consumer responses to their products, but also serve as a reliable source of information help other customers make purchase decision. In this paper, we propose a novel semi-supervised learning algorithm to address the feature-level reviews mining problem. The proposed method consists of three phases: (1) build a support function that characterizes the support of a multi-dimensional distribution of a given data set; (2) decompose a whole data space into a small number of separate clustered regions via a dynamical system associated with the constructed support function; (3) assign a class label to each decomposed region using the information of their constituent labeled data and the constructed dynamical system, thereby classifying in-sample unlabeled data as well as unknown out-of-sample data.

Keywords: Online reviews; Text mining; Semi-supervised learning; Support vector machine (SVM) (search for similar items in EconPapers)
Date: 2015
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-662-43871-8_102

Ordering information: This item can be ordered from
http://www.springer.com/9783662438718

DOI: 10.1007/978-3-662-43871-8_102

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().