Multi-objective bat algorithm for mining numerical association rules

Numerical association rule mining problem attracts the attention of researchers because of the various applications and its importance in our world with the fast growth of the stored data. ARM is computationally very expensive because the number of rules grows exponentially as the number of items in the database increases. Generally, ARM is more complex when we introduce the quality criteria and the usefulness to the user. In this paper deals with the problem of numerical ARM. In which, we propose a new multi-objective meta-heuristic called multi-objective bat algorithm for association rules mining (MOB-ARM). To identify more useful and understandable rules, we introduce four quality measures of association rules: Support, confidence, comprehensibility, and interestingness, in two objective functions. A series of experiments are carried out on several well-known benchmarks in ARM field and the performance of our proposal are evaluated and compared with those of other recently published methods including mono-objective and multi-objective approaches. Also, the paper presents a comparative study with three other methods dealing with multi-objective association rule mining. The obtained results show that our method is competitive with other methods and extract useful and understandable rules.