Service quality evaluation and service improvement using online reviews: A framework combining deep learning with a hierarchical service quality model

In the era of big data, service quality evaluation using online reviews has become a popular topic. However, very few studies focus simultaneously on service quality evaluation and service improvement. In this study, a research framework for service quality evaluation and service improvement is proposed, sentiment analysis is used to extract the temporal scores of the service attributes of each subdimension of the service quality model from online reviews, and a long short-term memory network is used to predict the scores for the service quality provider. Furthermore, a long short-term memory network-based sensitivity analysis, in conjunction with improvement costs, is used to rank the subdimensions in the service quality model. Then, service improvement strategies are determined according to the rankings of the service attributes. Hotels’ online reviews were used to investigate the effectiveness of the proposed framework. A series of service improvement strategies for the specific service attributes are provided.