Recommender systems, as one of well-known Web intelligence applications, aim to alleviate the information overload problem and produce item suggestions tailored to user preferences. Typically, user preferences or tastes are collected through users’ implicit or explicit feedback in various formats, such as user ratings, online behaviors, text reviews, etc. Also, user feedback on different items can be collected from several systems or domains. The diversity of feedback formats and domains provides multiple views to users’ preferences, and thus, can be helpful in recommending more related items to users. Cross-domain recommender systems and transfer learning approaches propose to take advantage of such diversity of viewpoints to provide better-quality recommendations and resolve issues such as the cold-start problem.
The emerging research on cross-domain, context-aware and multi-criteria recommender systems, has proved to be successful. Given the recent availability of cross-domain datasets and novelty of the topic, we organize the 1st workshop on intelligent recommender systems by knowledge transfer and learning (RecSysKTL) held in conjunction with the 11th ACM Conference on Recommender Systems. This workshop intends to create a medium to generate more practical and efficient predictive models or recommendation approaches by leveraging user feedbacks or preferences from multiple domains. This workshop will be beneficial for both researchers in academia and data scientists in industry to explore and discuss different definition of domains, interesting applications, novel predictive models or recommendation approaches to serve the knowledge transfer and learning from one domain to another.
The definition of “domain” may vary in different applications, e.g., it could be (but not limited to):
- From one application to another: We may utilize user behaviors on social networks to predict their preferences on movies (e.g., Netflix, Youtube) or music (e.g., Pandora, Spotify).
- From one category to another: We may predict a user’s taste on electronics by using his or her preference history on books based on the data collected from Amazon.com.
- From one context to another: We may collect a user’s preferences on the items over different time segment (e.g., weekend or weekday) and predict her preferences on movie watching within another context (e.g., companion and location).
- From one task to another: It may be useful for us to predict how a user will select hotels for his or her vocations by learning from how he or she books the tickets for transportations.
- From one structure to another: It could be also possible for us to infer social connections by learning from the structure of heterogeneous information neworks.