A business context aware decision-making approach for selecting the most appropriate sentiment analysis technique in e-marketing situations

In the era of social media, consumers interact and exchange information, subjective opinions, feelings or thoughts about a given issue, a product or a service, via Web 2.0 technologies. As a result, electronic Word of Mouth, eWOM, has gained increasing influence on the formation of user opinions. Therefore, extracting and analyzing eWOM statements by means of Sentiment Analysis techniques is currently a hot topic that could potentially provide companies with key information about their customers. However, among all the available techniques, choosing the most appropriate one according to the analysis criteria is a real challenge. In this contribution, we first study the main desirable criteria for a sentiment analysis approach and classify the different existing approaches based on these criteria to try and provide a solution to this issue. We then propose a Multi-criteria Decision-Making methodology that selects the most suitable technique for each business case by taking their particular criteria into account.