I'm using association rules for a project and noticed that there is a dearth of papers in the last 10-15 years on the topic although it seemed really popular 15-25 years ago. Is there any reason why?
Association rules are a relatively straightforward class of algorithms. Those earlier papers cover most of the interesting properties of association rules.
The field of recommender systems moved towards collaborative filtering and matrix factorization around that time. Those methods have increased empirical performance and are more interesting to study.