Discovering Hidden Concepts in Predictive Models for Texts' Polarization

Discovering Hidden Concepts in Predictive Models for Texts' Polarization

Caterina Liberati (Department of Economics, Management and Statistics, University of Milano-Bicocca, Milan, Italy) and Furio Camillo (Department of Statistics, University of Bologna, Bologna, Italy)
Copyright: © 2015 |Pages: 20
DOI: 10.4018/ijdwm.2015100102
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The growth of Internet and the information technology has generated big changes in subjects' communication, which, nowadays, occurs through social media or via thematic forums. This challenges the traditional notion of Customer Relationship Management (CRM) and pushes businesses to prompt and accurate understanding of sentiments expressed, in order to address their marketing actions. In this paper, the authors propose a combined application of a supervised Sentiment Analysis (SA) with a probabilistic kernel discriminant to provide a robust classifier of texts polarization. The partition obtained is also described by means of a statistical characterization of the texts. Such an approach is very promising, not only in terms of classification accuracy, but also in terms of knowledge extraction. A real case study is illustrated in order to test and show the effectiveness of the proposed strategy.
Article Preview

Introduction

The importance of word-of-mouth has been widely stated in a large part of the marketing literature (Anderson, 1988; Goldenberg, Libai & Muller, 2001). The concept is always been related to the perceptions of a product/service. Anderson (1998) stated that unsatisfied customers do engage in greater word-of-mouth than satisfied ones. Similarly, Goldenberg, Libai & Muller (2001) claimed that consumers’ decision-making process is strongly influenced by word-of-mouth. As such, word-of-mouth communications at the micro-level can influence macro-level phenomena (Ye, Zhang & Law, 2009). Recently, the unprecedented growth of Internet and the information technology has generated big changes in subjects’ communication that, more and more frequently, occurs through social media or via thematic forums. This has resulted in a surge of information that is freely available online in a text format. For example, many online forums and review sites exist for people to post their opinions about a product (Coussement & Van den Poel, 2008). While such consumer-generated content offers possibility to businesses to evaluate their credibility and to monitor the mood of their markets, some concerns arise, relative to how to manage such new interaction between customer and firm. On one hand, this challenges the traditional notion of Customer Relationship Management (CRM), where the company is the main actor, addressing passive customers, whose ability to respond to the company's efforts is essentially captured in their purchase behavior (Malthouse, Haenlein, Skiera, Wege & Zhang, 2013). With the rise of vast social networking platforms, the customer is no longer limited to a passive role in his relationship with a company so the transition from a traditional view of CRM to a social-CRM perspective seems necessary (Greenberg, 2009; Greenberg, 2010; Malthouse, Haenlein, Skiera, Wege & Zhang, 2013). On the other hand, the web 2.0 represents an opportunity: prompt and accurate understanding of sentiments expressed within the online text could lead to effective marketing strategies and service recovery and could help the management to identify key drivers for improved customer relationships. Therefore, the main issues, here, are twofold: first, how to transform the Voice of Customer (VoC), i.e. such unstructured text data into analyzable structured data; second issue is how to build a semi-automatic process for classifying and describing customer opinions, in order to provide useful segmentation for marketing actions. Thus, addressing both questions means combine a parsing and filtering analysis with an effective supervised classification easy-to-read in terms of relevant original texts.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 13: 4 Issues (2017)
Volume 12: 4 Issues (2016)
Volume 11: 4 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing