HW5_2010 - as mentioned on p. 7-12: LOC=rejected. With the...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
DSCI 4520/5240 – Data-Based Decision Support Systems HW5 Data: PROSPECT Analysis: K -means Clustering Due: Nov 10 HW Instructions: 1. Open the PROSPECT data using SAS Enterprise Miner 5.3 2. Follow the analysis steps described on the handout distributed in class. 3. Turn in a printed word document with (i) a brief one-paragraph description of the PROSPECT problem, (ii) a brief one-paragraph outline of your analysis including a screen shot of your final Diagram, (iii) screen shots of various analysis steps as listed below, and (iv) your answer to the 2 questions listed at the end. 5.1. Build the analysis diagram shown on your handout. Remember to set the model role
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: as mentioned on p. 7-12: LOC=rejected. With the Impute node, use Tree for both class and interval variables. 5.2. Select the Clustering node and select Std. Deviation as the standardization method. Set the selection method to User Specify. Change the maximum number of clusters to 4 . Inspect the Segment plots that are shown in the results ( Screen #1 ). 5.3. Maximize the Mean Statistics window shown in the results. Transfer it to your report ( Screen #2 ). Questions: 1. When we originally started setting up the clustering node, why did we have to standardize the data? 2. Provide an interpretation of the clustering solution with the 4 clusters....
View Full Document

Ask a homework question - tutors are online