Based On Data Mining Technology Customer Churn Analysis |
Posted on:2005-10-20 | Degree:Master | Type:Thesis |
Country:China | Candidate:K Yang | Full Text:PDF |
GTID:2208360122475581 | Subject:Computer application technology |
Abstract/Summary: | PDF Full Text Request |
Data mining is the extraction of patterns representing valuable knowledge implicitly stored in large databases or data warehouses. This paper introduces how the data mining technology apply in the prediction of customer churn. The author takes CRISP-DM as the referenced model of the data mining process. In the execution process of data mining, the author reduces the dimensions with the method of neural network learning and produce rule sets with the method of decision tree learning. The resulting model is improved not only on the speed of training but also on the classification precision and intelligibility. Lastly the paper discusses how to develop data mining applications with the model and gives the realization in a real project. |
Keywords/Search Tags: | Neural network, Decision tree, Data mining, Classification |
PDF Full Text Request |
Related items |