DNSC 6279 Data Mining Fall 2017

Assignment 3 Due: Tuesday, Nov 14- 11:59 PM ET

-----------------------------------------------------------------------

Please use blackboard to submit your assignment (click on Assignment 3, browse to attach your files
and click the submit button). HARD COPIES OR EMAIL SUBMISSIONS WILL NOT BE ACCEPTED!

You need to submit a final written report. In your report make sure all the questions are answered and
addressed explicitly, and also add your critical evaluation of the situation and the analysis method used
to answer these questions.

The final report should be uploaded in the PDF format. (If you don't have a PDF printer installed on
your computer, you can search for a free word to PDF converter online).

You should also submit all your supporting files

You are only allowed a SINGLE attempt to upload your files. So upload them only after you finalize
your solutions. Email the TA if you have any issues doing so.
 Question 1: Association Analysis

A drug store chain wants to learn more about cosmetics buyers purchase patterns. Specifically, they want to know
what items are purchased in conjunction with each other, for purposes of display, point of sale special offers, and to
eventually implement a real time recommender system to cross-sell items at time of purchase. The data (in the file
Cosmetics.jmp) are in the form of a matrix in which each column represents a product group, and each row a
customer.

Conduct an association analysis of the data set, and identify five rules that you would deem interesting and/or useful.
For the rules that you identify, explain how the reported support, confidence, and lift values are calculated.

Question 2: Prediction using Neural Nets

Car Sales. Consider again the data on used cars (ToyotaCorolla.jmp) with 1436 records and details on 38 attributes,
including Price, Age, KM, HP, and other specifications. The goal is to predict the price of a used Toyota
Corolla based on its specifications.

a. Determine which variables to include, and use the neural platform in JMP Pro to fit a model. Use the
validation column for validation, and use the default values in the Neural model launch dialog. Record the
RMSE for the training data and the validation data, and save the formula for the model to the data table
(use the Save Fast Formulas option, which will save the formula as one column in the data table). Repeat
the process, changing the number of nodes (and only this) to 5, 10, and 25.

i. Using your recorded values, what happens to the RMSE for the training data as the number of nodes
increases?

ii. What happens to the RMSE for the validation data?

iii. Comment on the appropriate number of nodes for the model.

iv. Use the Model Comparison platform to compare these four models (use the Validation column as either
a By variable or as a Group variable, and focus only on the validation data). Here, RASE is reported
rather than RMSE. Compare RASE and AAE (average absolute error) values for these four models.
Which model has the lowest error?

b. Conduct a similar experiment to assess the effect of changing the number of layers in the network as well as
the activation functions.