Consumer Online Behavior Analysis and CRISP-DM Process Model Development

Abstract

Understanding the behavior of a company's customers forms the cornerstone of modern marketing. Data analysis is penetrated by cutting-edge artificial intelligence techniques like data mining and machine learning. These techniques can be used in a variety of businesses and for online sales of any type of commodity in big volumes. They are frequently employed in the sale of clothing, computers, and electronics. However, they can also be used in the B2B as well as B2C sales of seeds, agricultural goods, or agricultural machinery in the agricultural sector. A selling company might gain competitive advantages or higher profits by combining the right offers and customer knowledge. Our study's major objective is to create a CRISP-DM process model that will let small firms examine the behavior of their online clients. We analyze the online sales data using machine learning techniques including clustering, decision trees, and association rules mining to achieve the main objective. The usage of the proposed model in the area of online sales in the agriculture sector is discussed after the proposed model has been evaluated.