The video discusses mining various kinds of association rules in data mining. It explains different types of association rules such as multi-dimensional, multi-level, quantitative, and statistical. The focus is on mining multi-level association rules which involve mining rules at multiple levels of abstraction with the ability to move quickly between levels. This is done by setting a uniform minimum support for all levels and using algorithms like Apriori to eliminate data items with support less than the minimum. The goal is to effectively mine data at different levels of abstraction by supporting a consistent minimum level of support across all levels.