Improved Classification Approach for Business Intelligence‎Using Data Specific and Feature Oriented Min-Max Normalization

Business intelligence is becoming more and more important to managers in today's corporate setting as a crucial aspect of information ‎technology. Many companies are eager to use intelligent technologies to enhance their decision-making processes in corporate operations. ‎Accordingly, intelligence is typically linked to human-like traits, including the ability to assimilate new information, learn from mistakes, and ‎pursue objectives similar to those of a human. One of the main objectives of business intelligence integration across businesses is to ‎generate reports through management dashboards that use key indicators to facilitate informed decision-making. The objective of this paper is to ‎develop a practical model that uses machine learning indicators and algorithms to optimize the product sales system using classification ‎techniques. Several factors are used by the model to enhance client classification techniques. Additionally, the study uses association rules to ‎look at clients' shopping carts, find links between the items they have bought, and generate personalized offers based on the rules they have ‎discovered. Different classification techniques, including C4.5, random forest, and reduced error pruning tree, are used to compare with the ‎suggested methodology before being used to evaluate and improve the results. The findings show that the best results should be obtained ‎using the recommended methodology in conjunction with the indicated model‎.