Assignment 1
Assignment 2
273 Business Intelligence for Analytical Decisions
This assignment must be completed individually. Please submit assignment to Canvas dropbox
including your name and student ID number.
1. You have seen how nominal and numeric attributes are specified in an ARFF file. Consider a
feature called NUM_OF_DEPENDENT, which in the data file can take any value from the
set {0,1,2,3}.
In the ARFF file, this feature could be specified as:
@attribute DEPENDENT Numeric
or as:
@attribute DEPENDENT {0,1,2,3}
What is the difference between the 2 specifications? Which specification would you use and
why?
1 point
2. For this practice set, you should be working with the data set called affiliation. This data set
includes votes for each of the U.S. House of Representatives Congressmen on the 16 key
votes identified as 16 different attributes in the data set. We are going to use a portion of this
data set (marked for training) to train our classification models. The goal of the classification
is simple: given the stands (votes) of an individual congressman, can we predict his/her party
affiliation? As to be expected with any real-world dataset, there are several records with
NULL values. However, we have taken a subset of the dataset with only those records that
do not have a NULL value. The table training-no-NULL should be used to train (build)
classification models, and the table testing-no-NULL should be used to test the results. The
appropriate ARFF files are provided at EEE. You can use Excel for parts (a), (b) and (c).
Use Weka for part (d).
(a) Prepare a frequency (count) chart for the data set and populate it based on the training data.
See examples covered in class. Frequency chart shows cross-tab of class variables with each
of the other variables. Use Excel Pivot tables for this – you may need several pivot tables.
(b) Prepare a populated probability chart (conditional probability) from the frequency chart in
part (a). Again see examples covered in class.
(c) Based on the probability chart, apply naïve Bayesian classification to the test cases in the data
set (5 rows in testing-no-NULL.ARFF). Compare your classification result with the actual
data. What is the percentage of accurate classification from the classifier?
(d) Use WEKA to run the Naïve Bayes classifier on training-no-NULL.ARFF and set test file as
testing-no-NULL.ARFF. Report the confusion matrix output by Weka.
7 points
3. After running a classifier in WEKA on some dataset, the following confusion matrix was
obtained:
==========Confusion Matrix=========
a b ← classified as
921 28 |a=yes
17 374 |b=no
(a) Based on this confusion matrix, estimate the overall accuracy of the classifier.
(b) Estimate the stratified accuracies of the classifier.
2 points