Project 8: Applied Theory & Practice I, Mushrooms

Objectives

• Demonstrate your understanding of the data mining process
• Learn about the Knime and Weka data mining toolkits.
• Discover patterns in the mushroom dataset

Requirements

Determine what patterns exist in a dataset containing attributes of a certain mushroom species. Apply clustering, classification, anomaly detection and association analysis to answer the following questions.

• Can you generate summary statistics that help describe the data?
• Can the edible and poisonous data objects be distilled into groups?
• Can a classification model be created that can predict whether a mushroom is edible or poisonous?
• Do any anomalies exist in the dataset?
• Can any association rules be generated from this dataset?

The dataset is provided to you in csv format here and some metadata is here.

This project is exploratory and may require you to preprocess the data to "fit" the algorithms your wish to apply. You should explore the data, and use scripts, Weka and Knime to generate qualified conclusions to the questions above.

You must create a brief paper that answers the questions listed above, and what process you undertook to reach your conclusions.

Grading Criteria (1000 points)

I must be able to pull your repository which contains your writeup of your data mining process.

Due Date

This assignment is due by midnight on Monday, November 21.