Assessing reliability reputation of products based on online customer reviews

The traditional concept of functional reliability refers to the ability of an item to perform a required function without failure, under stated conditions for a stated period of time. Numerical indexes of functional reliability are estimated from laboratory test data or field failure data. Theses indexes are not always available to the customers. On the other hand, reliability reputation, which concerns the opinions of the general public regarding the functional performance of a given product, is dominant in affecting the purchase willingness of the customers. In this paper, we develop a new method for evaluating the reliability reputation of a given product. Numerical metrics are first defined to measure reliability reputation. Online customer's reviews are collected and used as data for reliability reputation assessment. Text mining algorithms are used to extract information regarding the reliability of the product and the maintenance service of the company. Finally, an integrated approach is proposed to evaluate the reliability reputation. A case study is conducted to demonstrate the validity of the developed methods.