An Integrated Text Analytic Framework for Product Defect Discovery

Abstract

T he recent surge in the usage of social media has created an enormous amount of user-generated content (UGC). While there are streams of research that seek to mine UGC, these research studies seldom tackle analysis of this textual content from a quality management perspective. In this study, we synthesize existing research studies on text mining and propose an integrated text analytic framework for product defect discovery. The framework effectively leverages rich social media content and quantifies the text using various automatically extracted signal cues. These extracted signal cues can then be used as modeling inputs for product defect discovery. We showcase the usefulness of the framework by performing product defect discovery using UGC in both the automotive and the consumer electronics domains. We use principal component analysis and logistic regression to produce a multivariate explanatory analysis relating defects to quantitative measures derived from text. For our samples, we find that a selection of distinctive terms, product features, and semantic factors are strong indicators of defects, whereas stylistic, social, and sentiment features are not. For high sales volume products, we demonstrate that significant corporate value is derivable from a reduction in defect discovery time and consequently defective product units in circulation.

A Domain Oriented LDA Model for Mining Product Defects from Online Customer Reviews

“…Our analysis is consistent with existing studies of product defect discovery in that extracting product defects from online reviews has important business value [3], [32], [33]. Our results also indicate that the product defect information is not one-dimensional but draws on values associated with domain-oriented attributes (e.g.…”

“…Taking product defect disclosure text analysis for example, researchers [32] create a distinctive key word list named as "smoke words" which appear significantly more frequent in vehicle defects than other postings. Supervised learning methods [21] provide another method for summarizing documents to predefined categories. For example, some key terms, product features, and semantic factors can help identify product defects, but stylistic, social, and sentiment features cannot [3].…”

“…Most of early works on latent aspect-based opinion analysis are based on the frequency of noun phrases to identify aspects [16]. Later works are based on some dictionary methods or supervised learning techniques to learn aspects and their ratings [21]. However, most of current studies [22] are based on unsupervised topic models or LDA.…”

“…As described by [3], the business value of product quality is associated with product competitive advantage, and then, product success and commercial success. Here, we focus on the effects of product quality issues from user reviews on customer relationship and defect management.…”

“…Many companies seek to collect data on customer satisfactions and needs via survey, email, and work logs [3]. Motivated by customers' extrinsic benefits and intrinsic demands, they would also like to share their experience on products in their feedbacks, which are helpful to other consumers to make purchase decisions and product managers for improving their service and product quality.…”