A Proposed Model for Multi-Dimensional Data Quality‏ ‎

نوع المستند : المقالة الأصلية

المؤلفون

1 Data Analyst and Officer Administrator, International Ranking Unit, Helwan ‎University, Cairo, Egypt.‎

2 Information Systems Department, Faculty of Commerce and Business Administration, Helwan University, Cairo, Egypt.

المستخلص

In recent years, business intelligence has emerged as a critical ‎field leveraging data analysis to generate actionable insights for ‎informed decision-making.‎‏ ‏This paper emphasizes the ‎significance of data quality and the choice of the most ‎appropriate model to enhance the accuracy of predictions,‎‏ ‏which ‎contributes to improving marketing strategies and banking ‎decision-‎making. The aim of this paper is to evaluate the ability ‎of machine learning models to predict the outcomes of direct ‎‎marketing campaigns for banks using accurate and unbiased ‎data. The five models tested were K-Nearest Neighbour's ‎‎‎(KNN), Random Forest, Decision Tree, Gradient Boosting, and ‎Multi-Layer Perceptron (MLP). The data showed that ‎the ‎Gradient Boosting model performed better than others in ‎marketing applications, with an accuracy of 91.45%. ‎Random ‎Forest showed similar performance with an accuracy of 91.24% ‎despite the longer prediction time, (MLP) ‎achieved 90.24%, ‎‎(KNN) achieved 89.55%, and Decision Tree was the fastest, but ‎its accuracy was somewhat lower ‎at 88.58%.‎

الكلمات الرئيسية

ملفات

شارك

إرسال الاستشهاد إلى

الإحصائيات