Classification Theorem : Classification Algorithm

1419 Words Mar 9th, 2015 6 Pages
J48 Classification Algorithm
The J48 classification algorithm considers all the possible tests that can split the data set and selects a test that gives the best information gain. So, whenever it encounters a training set it identifies the attribute that discriminates the various instances in a most clear way. From the possible values, if there is an attribute with unique value and if it has the same value as that of target value then the branch is ended For the other cases, other attribute with the highest gain is considered. This process is continued till we obtain a combination of attributes with the highest information gain or till we do not have any more attributes to test. The scenario in which we do not have any more attributes, the branch is assigned an average value or the value majorly possessed by the items.

One Rule Classification Algorithm
OneR also known as One Rule is a classification algorithm that generates one rule for each predictor in the data set. The attributes used for classification are known as predictors and the attribute classified is known as the target. After the generation of rule, the rule with the smallest error is considered as the “one rule.” A frequency table is generated for the creation of rule for a predictor. The results generated with the help of oneR algorithm are easy to interpret and are in easy readable format.
Algorithm:
For each predictor, For each value of that predictor, evaluate rule by following these steps;…
Open Document