Reduction of dynamic data in fuzzy association rule mining

The toolbox is capable of performing incremental data reduction in fuzzy association rule mining.

This submission is to support the submission of the corresponding manuscript and can be briefly described as follows: incrementality is a property of a Data Mining technique/approach to tune its results to some novel data records without having to reprocess the whole dataset again, and this submission provides an incremental version of the data reduction technique proposed in https://doi.org/10.1016/j.eswa.2020.113781.

In general, scripts should be launched in the following order:

1. Subset_Design.m

INFO: A dataset can be divided into several subsets.

2. NonIncremental_Data_Reduction.m or Incremental_Data_Reduction.m

INFO: (a) This step is optional, since the following steps can also be perfomed on not reduced data. (b) In order to perform incremental reduction, at least one non-incremental reduction process has to be performed on the data before getting additional incrementally-processed records. (c) Results of data reduction get saved into “Data_KM_Final_Set.txt”.

3. Data_Preparation_and_Partitioning.m

INFO: One selects a dataset to be put through ARM, defined partitioning and explicitly defines antecedent and consequent attributes.

4. Show_Partitioning.m

INFO: The defined partitioning is shown in a graphical form.

5. ARM.m

INFO: This script performs association rule mining (ARM).

6. Classifier_Design.m

INFO: This script creates an associative classifier.

7. Classifier_Running.m

INFO: This script runs the classifier on some testing data.