Recently, a few papers report counter-intuitive observations made from experiments on recommender system (RecSys). One observation is that users who spend more time and users who have many interactions with a recommendation system receive poorer recommendations. Another observation is that models trained by using only the more recent parts of a dataset show significant performance improvement. In this opinion paper, we interpret these counter-intuitive observations from two perspectives. First, the observations are made with respect to the global timeline of user-item interactions. Second, the observations are considered counter-intuitive because they contradict our expectation on a recommender: the more interactions a user has, the higher ...
Recommender systems are an increasingly important technology and researchers have recently argued fo...
In many life situations, people choose sequentially between repeating a past action in expectation o...
Part 1: ConferenceInternational audienceRecommender systems (RS) are designed to assist users by rec...
Datasets used for the offline evaluation of recommender systems are collected through user interacti...
In academic research, recommender systems are often evaluated on benchmark datasets, without much co...
Recommender models are hard to evaluate, particularly under offline setting. In this paper, we provi...
Recommendation systems are often evaluated based on user’s interactions that were collected from an ...
Recommender systems are ubiquitous in most of our interactions in the current digital world. Whether...
Recommender models are hard to evaluate, particularly under offline setting. In this paper, we prov...
Recommender systems daily influence our decisions on the Internet. While considerable attention has ...
Popularity is often included in experimental evaluation to provide a reference performance for a rec...
Numerous recommendation approaches are in use today. However, comparing their effectiveness is a cha...
Multi-objective recommender systems (MORS) provide suggestions to users according to multiple (and p...
Recommender systems learn from historical user-item interactions to identify preferred items for tar...
Recommender systems help people find relevant content in a personalized way. One main promise of suc...
Recommender systems are an increasingly important technology and researchers have recently argued fo...
In many life situations, people choose sequentially between repeating a past action in expectation o...
Part 1: ConferenceInternational audienceRecommender systems (RS) are designed to assist users by rec...
Datasets used for the offline evaluation of recommender systems are collected through user interacti...
In academic research, recommender systems are often evaluated on benchmark datasets, without much co...
Recommender models are hard to evaluate, particularly under offline setting. In this paper, we provi...
Recommendation systems are often evaluated based on user’s interactions that were collected from an ...
Recommender systems are ubiquitous in most of our interactions in the current digital world. Whether...
Recommender models are hard to evaluate, particularly under offline setting. In this paper, we prov...
Recommender systems daily influence our decisions on the Internet. While considerable attention has ...
Popularity is often included in experimental evaluation to provide a reference performance for a rec...
Numerous recommendation approaches are in use today. However, comparing their effectiveness is a cha...
Multi-objective recommender systems (MORS) provide suggestions to users according to multiple (and p...
Recommender systems learn from historical user-item interactions to identify preferred items for tar...
Recommender systems help people find relevant content in a personalized way. One main promise of suc...
Recommender systems are an increasingly important technology and researchers have recently argued fo...
In many life situations, people choose sequentially between repeating a past action in expectation o...
Part 1: ConferenceInternational audienceRecommender systems (RS) are designed to assist users by rec...