Efficient Mining of High Confidience Association Rules without Support Thresholds

摘要: Association rules describe the degree of dependence between items in transactional datasets by their confidences. In this paper, we first introduce problem mining top rules, namely those association with 100% confidence. Traditional approaches to need a minimum support (minsup) threshold and then can discover supports ≥ minsup; such approaches, however, rely on minsup help avoid examining too many candidates they miss whose are below minsup. The low (e.g. some unusual combinations factors that have always caused disease) may be very interesting. Fundamentally different from previous work, our proposed method uses dataset partitioning technique two border-based algorithms efficiently all given consequent, without constraint threshold. Importantly, use borders concisely represent instead enumerating them individually. We also discuss how zero-confidence high (say 90%) confidence using similar rules. Experimental results Mushroom, Cleveland heart disease, Boston housing reported evaluate efficiency approach.