Abstract

The case presented in this paper describes an early prototype and next steps for developing a user-adaptive recommender system using semantic analysis and matching of user profiles and content. Machine learning methods optimize semantic analysis and matching based on implicit and explicit feedback of users. The constant interaction with users provides a valuable data source that is used to improve human-computer interaction and for adapting to specific user preferences. This can lead to, among others, higher accuracy and relevance in content matching, more intuitive graphical user interfaces, improved system performance, and better prioritization of tasks.

Share

COinS
 

System Learning of User Interactions

The case presented in this paper describes an early prototype and next steps for developing a user-adaptive recommender system using semantic analysis and matching of user profiles and content. Machine learning methods optimize semantic analysis and matching based on implicit and explicit feedback of users. The constant interaction with users provides a valuable data source that is used to improve human-computer interaction and for adapting to specific user preferences. This can lead to, among others, higher accuracy and relevance in content matching, more intuitive graphical user interfaces, improved system performance, and better prioritization of tasks.