Optimized fuzzy association rule mining for quantitative data

Abstract

With the advance of computing and electronic technology, quantitative data, for example, continuous data (i.e., sequences of floating point numbers), become vital and have wide applications, such as for analysis of sensor data streams and financial data streams. However, existing association rule mining generally discover association rules from discrete variables, such as boolean data ('0' and '1') and categorical data ('sunny', 'cloudy', 'rainy', etc.) but very few deal with quantitative data. In this paper, a novel optimized fuzzy association rule mining (OFARM) method is proposed to mine association rules from quantitative data. The advantages of the proposed algorithm are in three folds: 1) propose a novel method to add the smoothness and flexibility of membership function for fuzzy sets; 2) optimize the fuzzy sets and their partition points with multiple objective functions after categorizing the quantitative data; and 3) design a two-level iteration to filter frequent-item-sets and fuzzy association-rules. The new method is verified by three different data sets, and the results have demonstrated the effectiveness and potentials of the developed scheme.

Dynamic optimisation based fuzzy association rule mining method

“…The first subsection explains the corresponding methods, parameters and datasets. The second subsection lists the antecedents of strong rules and the results of partitioning points which indicate our with our previous work OFARM method [16] with data analysis of efficiency, as the method GFARM lose the general comparison conditions (details will be explained in this section). With all of the experimental studies, we can simply further represent the benefits of our DOFARM method, including the good performance of efficiency, effectiveness and accuracy expecting theoretical demonstrations in section 3.…”

“…In our experiment, the proposed DOFARM method is evaluated by comparing with GFARM method [3] and OFARM method of our previous work [16]. From the experimental descriptions among this section, we see our novel DOFARM method extends GFRAM and OFARM method to arbitrary parameters and metrics and improves it on accuracy, effectiveness and efficiency.…”

“…In this subsection, we will introduce our scheme to optimise metrics of fuzzy association rules simultaneously which is based on Richardson Extrapolation and Gradientbased optimisation methods [16], [17], [18], [19], [20], [21]. A theorem is illustrated to indicate the correctness of our multi-objective optimisation process.…”

“…. , DP 4 that is applied to distinguish different intervals is computed for a given continuous feature as described in paper [16]; 2: for ∀X components x 0 , x 1 in every continuous feature do 3:…”

Mineração de regras de associação temporais quantitativas por meio de algoritmo genético

“…Nosúltimos anos, vários métodos de mineração de dados temporais têm sido propostos [1,2,3,4,5,7,10,11,15,17,18,19], assim como de mineração de dados quantitativos [8,6,12,13,14,20]. Contudo, os métodos existentes na literatura não são efetivamente aplicáveis para a mineração de regras (implicações) de dados temporais quantitativos, devido a deficiências em satisfazer os seguintes critérios:…”