tailieunhanh - Collective Intelligence in Action phần 8

Tạo các thuộc tính. Tạo các tập dữ liệu cho việc học tập. Xây dựng các mô hình dự báo. Đánh giá chất lượng của mô hình xây dựng. Dự đoán số thông tin đăng nhập cho một người dùng tôi thực hiện một lớp WEKATutorial, sau những năm bước. Mã cho lớp này được hiển thị trong danh sách 7,1. | Classification fundamentals 275 Simpo PDF Merge and Split Unregistered Version - http apply the model to make predictions. The application of the mathematical model for predictions is typically fast and can be used for real-time predictions in an application while the amount of time taken to build the predictive model is much greater and is typically done asynchronously in the application. In this chapter we review some of the key supervised learning algorithms used for both classification and regression. We build on the example from the previous chapter of clustering blog entries. We use a simple example to illustrate the inner workings of the algorithms. We also demonstrate how to build classifiers and predictors by using the WEKA APIs. For this we apply the APIs to live blog entries retrieved from Technorati. Three commonly used classification algorithms are covered in this chapter decision trees Naive Bayes and Bayesian networks also known as belief networks or probabilistic networks . The key regression algorithms covered in this chapter include linear regression multi-layer perceptron and radial basis functions. We also briefly review the JDM APIs related to classification and regression. At the end of this chapter you should have a good understanding of the key classification and regression algorithms how they can be implemented using the WEKA APIs and the related key JDM concepts. Classification fundamentals In most applications content is typically categorized into segments or categories. For example data mining-related content could be categorized into clustering classification regression attribute importance and association rules. It s quite useful especially for user-generated content to build a classifier that can classify content into the various categories. For example you may want to automatically classify blog entries generated by users into one of the appropriate categories for the application. One common example of a .