SD4ft-Miner Demonstration - Paramsk

 

Example

In this part we will show an example of SD4ftTask. We will use patients' medical data of the STULONG study where there are 1419 objects (patients). This will be very simple example so we will use only 3 attributes:

Figure 1: Example

Figure 1: Input parameters

We solve the task Which groups of men differ greatly when we compare their height and weight. The difference of groups will be compared through the association quantifier Founded Implication.

Figure 1 shows our task in general. Here can be seen which attributes are set as Antecedent, Succedent, First set and Second set. No Condition is applied here.

As quantifiers are used Base FirstSet (which will ensure that number of men in the first set satisfying Antecedent and Succedent will be greater than the value given – 5), Base SecondSet (which will ensure that number of men in the second set satisfying Antecedent and Succedent will be greater than the value given – 5) and FUI DiffVal (which means that the difference between Founded implications of the both sets will be greater that the value given – 0.40).

Next four figures show the detail of the Antecedent, Succedent, First set and Second set. Here can be closely seen, how attributes are set. From each figure can be seen all necessary parameters as length of partial cedents and all parameters for each attribute.

Figure 2: Antecedent

Figure 2: Antecedent

Figure 3: Succedent

Figure 3: Succedent

Figure 4: First Set

Figure 4: First Set

Figure 5: Second Set

Figure 5: Second Set

After generating this task we get 3 hypotheses and one of them is displayed on fig. 6. This hypothesis says these things (if we analyze only the Founded implication quantifier:

Figure 6: Hypothesis

Figure 6: Hypothesis