On mining quantitative association rules from multi-relational data with FCA
Masahiro Nagao, Hirohisa Seki · 2016
We consider the problem of mining quantitative association rules (ARs) from a multi-relational database (MRDB), where a database contains multiple tables (relations), and attributes in a table are either categorical or numerical (or quantitative). To handle numerical data in a precise and efficient way, we consider (logical) conjunctions with interval constraints, using the notion of closed interval patterns (CIPs) proposed by Kaytoue et al. in FCA (Formal Concept Analysis). We then propose an algorithm for mining quantitative ARs which satisfy both a minimum support and a minimum confidence. We also propose a pruning method tailored to computing CIPs and show its correctness. We give some experimental results, which show the effectiveness of the proposed method, compared with the conventional methods such as a discretization-based approach or an optimization-based approach.