Research Article
Facilitating User Authorization from Imbalanced Data Logs of Credit Cards Using Artificial Intelligence
Table 1
Symbolic notations used in the proposed algorithm CtRUSBoost.
| | Dataset segment under consideration | | Hypothesis value obtained through support vector machine in iteration for the instance (this serves as a numeric confidence rating) | | Hypothesis value obtained through decision tree in iteration for the instance (this serves as a numeric confidence rating) | | Cumulative pseudo loss | | Parameter to update the weight factor | | Factor for normalizing the distribution of weights taking the full training dataset/or normalized value for the distribution | | Distribution of weights at iteration taking the full training dataset for the sample | | Distribution of weights at iteration taking the full training dataset | | Distribution of weights for temporary training dataset | | temporary training dataset | | row with values of all columns except the last one (i.e., label) | | A label for the row | | Minority class label | | Total number of iterations employed in the ML model | or | Total counts of samples present in the | | Rows/tuples in the dataset (excluding the last column having labeled entries) | | Total available labels in the dataset |
|
|