A comparison of services for intent and entity recognition for conversational recommender systems

Conversational Recommender Systems (CoRSs) are becoming increasingly popular. However, designing and developing a CoRS is a challenging task since it requires multi-disciplinary skills. Even though several third-party services are available for supporting the creation of a CoRS, a comparative study of these platforms for the specific recommendation task is not available yet. In this work, we focus our attention on two crucial steps of the Conversational Recommendation (CoR) process, namely Intent and Entity Recognition. We compared four of the most popular services, both commercial and open source. Furthermore, we proposed two custom-made solutions for Entity Recognition, whose aim is to overcome the limitations of the other services. Results are very interesting and give a clear picture of the strengths and weaknesses of each solution.