Identification of the to-be-improved product features based on online reviews for product redesign

Acquisition of customer needs usually serves as the basis for the identification of to-be-improved features for the product redesign process. However, the customer's true needs tend to be non-obvious and are difficult to extract from the data source like interviews or market survey. In the era of Big Data, with the advances in e-commerce, the customer's online review has become one of the most important data source to reveal the insight of customer's preference. In this paper, an online-review-based approach is introduced to identify the to-be-improved product features. The product features and corresponding opinions are extracted and reduced based on the semantic similarity. A structured preference model based on the semantic orientation analysis is constructed. A redesign index is subsequently introduced to measure the priority of redesign for each feature, and a target feature selection model is created to identify the to-be-improved features from candidate features considering engineering cost, redesign lead time and technical risk. A case study for smartphones is developed to demonstrate the effectiveness of the developed approach. In the future study, the online reviews may be combined with the traditional survey data to provide a more effective and reliable identification on the to-be-improved product features.