LAUSR

Big consumer opinion data provide valuable information about customer preferences. These online opinions facilitate designers to capture customer requirements (CRs) and understand customer satisfaction (CS) for new product development (NPD). A visible gap between practical significance and research studies is to uncover nonlinear relations between CRs and CS and derive strategic suggestions. Accordingly, a framework for online CRs Kano categorization is proposed. Firstly, both explicit and implicit features are extracted from opinionate texts to better capture CRs. Secondly, to evaluate the impact of CRs on the overall CS, a multi-layer neural network is invited, in which the impact from both positive and negative opinions over each product feature are distinguished. Finally, according to the estimated impact, CRs are categorized by a Kano Model based approach. To evaluate the effectiveness of the proposed framework, a case study that analyzes a large number of phone reviews is presented. Categories of studies were benchmarked to demonstrate the competitiveness of utilized approaches. This study is argued to disclose complex relations between CRs and the overall CS as well as strategic improvement suggestions by online opinion analysis. It enlightens designers to infer constructive strategies from big consumer opinion data for market-driven NPD.