Confusion matrix analysis of personal loan fraud detection using novel K means algorithm and linear regression algorithm.

This study aims to compare the performance of a K-means matrix analysis with that of a larger-sample linear regression approach in identifying instances of personal loan fraud. Tools and Techniques for Scientific Investigation: In order to see how well credit card analysis might detect fraudulent personal loans, we utilized an average accuracy rate of 10-80 samples. We assess how well a linear regression method and an innovative K-means strategy detect personal loan fraud in credit card records. A sample size of 10 is used to find the average accuracy of the two methods, and it is progressively increased to 80. End result: Results show that the K-means algorithm achieved an average accuracy rate of 92.75% with a standard deviation of 3.67849 and a standard error mean of 1.56109. In contrast, the linear regression approach has an average accuracy rate of 86.53%, a standard deviation of 1.56109, and a mean standard error of 0.63603. Evidently, the two sets are statistically different, as a p-value of 0.01 (P<0.05) suggests. T-test with two tails applied to distinct samples. The novel K-Means Method outperforms the Linear Regression Process in predicting the detection of personal loan fraud, according to this research. [ABSTRACT FROM AUTHOR]