Assignment 7:  Logic-based approaches

Use logic-based approaches (such as those listed under the group of rules and trees in Weka) to build classification models for the following two data sets, originating from the Irvine Machine Leaning Repository (http://archive.ics.uci.edu/ml/):

1.	Iris (http://archive.ics.uci.edu/ml/datasets/Iris)
2.	Congressional Voting Records (http://archive.ics.uci.edu/ml/datasets/Congressional+Voting+Records)

(For your convenience, I have placed both data sets in text format on this website.)

The models should help in classification of iris flowers (1) and congressmen (2) based on flower properties and voting record, respectively.  Please clean up and discretize the data, if needed (if you need domain knowledge, you can read more about the data sets in the descriptions placed at UCI Repository).  Make sure to try different learning algorithms and their parameters.  Once you have learned models, check and report their classification accuracy (by means of a cross-validation method, such as "leave-one-out").  Try to improve the accuracy by using different learning approaches, different values of parameters, and injecting outside knowledge.  Report the best classification accuracy that you have been able to achieve for each of the two data sets along with the methods that gave you the best accuracy.  Make sure to summarize your observations and conclusions.