Learning preferences of new users in recommender systems: an information theoretic approach

By Al Mamunur Rashid, George Karypis, John Riedl. 2008

267 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Management Science and Operations Research

Citations in the corpus

Learning Algorithms for Active Learning (2017, methodology): “In model-free strategies (Rashid et al., 2008), items are selected according to general empirical statistics such as popularity or informativeness.”
A collaborative filtering approach to mitigate the new user cold start problem (2012, background): “The new user problem [42,43,46] is among the great difficulties faced by the RS in operation.”
Recommender systems survey (2013, background): “The new user problem [190,197] represents one of the great difficulties faced by the RS in operation.”
Cold-start Active Learning with Robust Ordinal Matrix Factorization (2014, methodology): “Model-free strategies have been proposed for active data collection (Rashid et al., 2002; 2008), where empirical statistics of the data such as item popularity or rating entropy are used to select items.”
A survey of active learning in collaborative filtering recommender systems (2016, background): “• HELF [68]: stands for Harmonic mean of Entropy and Log- arithm of Frequency (HELF).”
Using Topic Models in Content-Based News Recommender Systems (2013, background): “Rashid et al. (2008) derive information theoretic strategies for how the ﬁrst recommendations for a new users should be made.”
Functional matrix factorizations for cold-start recommendation (2011, methodology): “…argued that an eﬀective model for designing the interview process should ask the interview questions adaptively, taking into account the user’s responses when asking the next question in the interview process, and decision trees have been found to work especially well for this purpose [7,19,20].”
Towards Deeper Graph Neural Networks with Differentiable Group Normalization (2020, background): “For example, in social networks, new users are often lack of profiles and tags [35].”
On bootstrapping recommender systems (2010, methodology): “A comprehensive study was conducted by the GroupLens team [8], and was later enhanced [9].”
Adaptive bootstrapping of recommender systems using decision trees (2011, methodology): “A comprehensive study was conducted by the GroupLens team [16], and was later enhanced [17].”
Interactive collaborative filtering (2013, methodology): “The items used in the interview can be selected based on measures such as popularity, entropy and coverage [30, 31], or a decision tree to partition the users [41, 17].”
A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems (2015, background): “The work in [18] studied how to learn user preferences for new users incrementally by recommending items that give the most information about user preferences while minimizing the probability of recommending irrelevant content.”
Current challenges and visions in music recommender systems research (2017, background): “Therefore, it is crucial to take into account a more realistic scenario when studying the active learning techniques in recommender systems,which can better picture how the system evolves over time when ratings are provided by users [143,146].”
Learning multiple-question decision trees for cold-start recommendation (2013, background): “This was also suggested by some experimental studies that show users prefer providing opinions on multiple items at a trial [17].”