CE802 Machine Learning And Data Mining

This document specifies a combined reassessment coursework assignment for CE802.Main aim of this assignment is to identify machine learning techniques appropriate for a particular practical problem and then describe the correct procedure for undertaking a comparative evaluation of several machine learning procedures when applied to the specific problem.It also assesses the general knowledge of the topics covered during the module.

Explain what is meant by the term information gain in the context of decision tree induction.

Describe two approaches to reducing the effect of overfitting in decision tree induction.

Describe the k-means clustering algorithm and state two of its limitations.

Explain the na ??ve Bayes method for classification and contrast it with a complete(or full)Bayes approach.

Explain the learning procedure known as bagging, paying particular attention to the procedure for generating training sets.

Describe how you would set about trying to assign the museum axes to the appropriate culture,given the information available. Your answer should include:

1. Discussion of the type of problem to be solved.
2. Selection of a small set of learning procedures, with an explanation of why they may be suitable.
3. A brief description of a comparative evaluation of the selected machine learning procedures.
4. Detailed description of how you would estimate the success of the final chosen system.
5. An account of how you would use the selected procedure to assign the museum axes to the appropriate culture.