No abstract

No abstract

No abstract

No abstract

Today, understanding customer satisfaction is becoming a difficult and complex task for companies due to the explosive growth of the voice of the customer in online reviews. This has pushed companies to rethink their business strategies and resort to business intelligence techniques in order to help them in analyzing customer requirements and market trends. This paper proposes a decision support framework for dynamically transforming the voice of the customer data into actionable insight. The framework measures the customer satisfaction by extracting key products’ aspects along with customers’ sentiments from online reviews using a text mining technique: the latent Dirichlet allocation approach. We apply the Fuzzy-Kano model to classify the real customer requirements, then, map them dynamically to the SWOT matrix. The proposed approach is extensively tested on an empirical dataset based on several performance metrics including accuracy, precision, recall, and F-score. The reported results showed that latent Dirichlet allocation approach has correctly extracted aspects with 97.4% accuracy and 92.4 % precision.