Optimizing Creditworthiness Assessment: Leveraging PCA and ESVM for Improved Credit Score Analysis

The paper proposes a novel approach to credit score analysis by combining Principal Component Analysis (PCA) with Ensemble Support Vector Machines (ESVM). This approach aims to address several limitations present in conventional credit score analysis techniques, including bias and fairness concerns, difficulties in handling unconventional credit data, lack of real-time data, inaccurate predictions, and unclear data. PCA is utilized to extract the most reliable and clear features from the dataset, which simplifies the classification process. By reducing the dimensionality of the data, PCA helps in improving the efficiency of the credit score analysis system. ESVM, a modified version of Support Vector Machines (SVM), is employed for classification tasks. SVM is known for its superiority in classification, and ESVM enhances its performance further by leveraging ensemble techniques. This combination enables accurate classification, thereby enhancing the accuracy of credit score predictions. To evaluate the proposed system, a credit scoring dataset is utilized in experiments. The results indicate that the introduced system outperforms conventional techniques in credit scoring analysis and offers better accuracy and reliability in assessing creditworthiness which providing financial institutions with better tools for risk management and decision-making.