A ve1y good reposito1y of data that have been used to test the performance of many machine-learning algorithms is available at ics.uci.edu/-mlearn/MLRepository.htmI. Some of t11e datasets are meant to test the lin1its of cu1Tent macl1inelearning algo1itl11Tis and to compare their performance wiili new approaches to learning. However, some of me smaller datasets can be useful for explo1i.ng the functionality of any data mining software or t11e software that is available as companion software with tl1is book, such as Statistica DataM.iner.

Download at least one dataset from iliis reposito1y (e .g., Credit Screening Databases, Housing Database) and apply decision tree or clustering metl10ds, as appropriate. Prepare a rep01t based on your results. (Some of t11ese exercises may even be proposed as semester-long projects for term papers.)