What is it about?
The paper introduces a method to profile a user's curiosity distribution. The curiosity model is called Probabilistic Curiosity Model, which models how broad or narrow a user's interest is, learnt from the user's previous accessed items. Breath means highly curious, narrow mean non-curious. Based on this personalized curiosity model, recommendations with the right novelty dosage that meets the user's novelty appetite are made.
Featured Image
Why is it important?
Every user is unique not only in what they are interested in but what in how much novelty and diversity they want from the items they read. The work is based on psychology's arousal theory, and recommendations are personalized according to the user's curiosity model, balanced with the recommendations' relevance to the user's interests.
Perspectives
This article applies well known findings in psychology to develop a computational method to estimate a user's curiosity model, based on which personalized recommendations are made that can balance relevance and novelty of the recommendations. The psychology part of this article is interesting to computer scientists. The findings are useful to psychologists working in human curiosity modeling because big data are used to verify a well known psychology curiosity model.
Dik Lun Lee
Read the Original
This page is a summary of: How Much Novelty is Relevant?, July 2016, ACM (Association for Computing Machinery),
DOI: 10.1145/2911451.2911488.
You can read the full text:
Contributors
The following have contributed to this page
