![]() | Only 14 pages are availabe for public view |
Abstract Recently, businesses have been growing, and becomes difficult for suppliers to know exactly what their clients need. Significant losses have encountered suppliers, as they often tend to purchase products that go unused. The lurking cause of this is the absence of knowledge and solid understanding of customer data. In efforts to study the problem and the data, a popular company that produces coffee, namely Yemeni Coffee, assembled information about each of its customers into a dataset, and made it available for employees to analyze. This research aims to use data mining and classification techniques on the client data, for two purposes. First, to understand the data by using visualization tools, such as Rapidminer to classify clients data and second to implement data-mining algorithms to predict the needs of each customer and hence, to recommend products for suppliers to purchase. The proper recommendation and efficiency of these algorithms depend greatly on the built model and the data set we analyzed. The project combines the benefits of feature selection and data mining classification techniques to accurately select and distinguish characteristics of clients geographic location, then consequently adduce a reliable model for an accurate proper recommendation for suppliers. The used datasets are described, results of our algorithms are presented and compared to other results published by other papers |