Rough Sets Data Analysis in Knowledge Discovery: A Case of Kuwaiti Diabetic Children Patients

Algorithm 1

RSBR discretization algorithm.

Input: Information system table (𝑆) with real valued attributes 𝐴_𝑖𝑗 and 𝑛 is the

     number of intervales for each attribute.

Output: Information table (𝑆𝑇) with discretized real valued attribute

1: for 𝐴_𝑖𝑗∈𝑆 do

2:  Define a set of boolean variables as follows: <span class="displayed-label" id="eq6">𝐵={^𝑛_𝑖=1𝐶_𝑎𝑖,^𝑛_𝑖=1𝐶^𝑛_{𝑏𝑖𝑖=1}𝐶_𝑐𝑖,…,^𝑛_𝑖=1𝐶_𝑛𝑖}(10)</span>

3: end for

 Where ∑^𝑛_𝑖=1𝐶_𝑎𝑖 correspond to a set of intervals defined on the variables of

 attributes 𝑎

4: Create a new information table 𝑆_new by using the set of intervals 𝐶_𝑎𝑖

5: Find the minimal subset of 𝐶_𝑎𝑖 that discerns all the objects in the decision class

 𝐷 using the following formula: <span class="displayed-label" id="eq7">Υ^𝑢=∧{Φ(𝑖,𝑗)∶𝑑(𝑥_𝑖≠𝑑(𝑥_𝑗))}(11)</span>

 Where Φ(𝑖,𝑗) is the number of minimal cuts that must be used to discern two

 different instances 𝑥_𝑖 and 𝑥_𝑗 in the information table.