代写COSC2110/COSC2111 Data Mining
100%原创包过,高质量代写&免费提供Turnitin报告--24小时客服QQ&微信：273427
代写COSC2110/COSC2111 Data MiningRMIT University
School of Computer Science and Information Technology
COSC2110/COSC2111 Data Mining
Assignment 1
This assignment counts for 23% of the total marks in this course.
Due Date 9:00am Monday 05 September 2016
PART 1: CLASSIFICATION 20 marks
1. This part of the assignment is concerend with the the file:
/KDrive/SEH/SCSIT/Students/Courses/COSC2111/DataMining/
data/other/bank-balanced.csv.
There is a description of the data in the file bank-names.txt in the same direc-
tory.
2. Run the following classifiers, with the default parameters, on this data: ZeroR,
OneR, J48, IBK and construct a table of the training and cross-validation errors.
You can get the training error by selecting “Use training set” as the test option.
What do you conclude from these results?
Run No Classifier Parameters Training Cross-valid Over-
Parameters Error Error Fitting
1 ZeroR None 30.0% 30.0% None
. . . . .
3. Using the J48 classifier, can you find a combination of the C and M parameter
values that minimizes the amount of overfitting? Include the results of your best
five runs, including the parameter values, in your table of results.
4. Reset J48 parameters to their default values. What is the effect of lowering the
number of examples in the training set? Include your runs in your table of re-
sults.
5. Using the IBk classifier, can you find the value of k that minimizes the amount
of overfitting? Include your runs in your table of results.
6. Try a number of other classifiers. Aside from ZeroR, which classifiers are best
and worst in terms of predictiveaccuracy? Include 5 runs in your table of results.
7. What are the implications of the above range of accuracies for developing a bank
application using classification techniques?
8. Compare the accuracy of ZeroR, OneR and J48. What do you conclude?
9. What golden nuggets did you find, if any?
10. [OPTIONAL] Use an attribute selection algorithm to get a reduced attribute set.
How does the accuracy on the reduced set compare with the accuracy on the full
set.
Data Mining 1 25-Jul-2016
Submit: Up to two pages that describe what you did for each of the above ques-
tions and your results and conclusions.
PART 2: NUMERIC PREDICTION 10 marks
1. Numeric Prediction of the Balance attribute in the bank data of part 1.
2. Run the following classifers, with default parameters, on this data: ZeroR, MP5,
IBk and construct a table of the training and cross-validation errors. You may
want to turn on “Output Predictions” to get a better sense of the magnitude of
the error on each example. What do you conclude from these results?
3. Explore different parameter settings for M5P and IBk. Which values give the
best performance in terms of predictive accuracy and overfitting. Include the
results of the best five runs in your table of results.
4. Investigate three other classifiers for numeric prediction and their associated pa-
rameters. Include your best five runs in your table of results. Which classifier
gives the best performance in terms of predictive accuracy and overfitting?
5. What golden nuggets did you find, if any?
Submit: Up to one page that describes what you did for each of the above ques-
tions and your results and conclusions.
PART 3: CLUSTERING 10 marks
1. Clustering of the bank data of part 1.
For this part use only the attributes Age, Marital, Education and Balance.
2. Run the Kmeans clustering algorithm on this data for the following values of K:
1,2,3,4,5,10,20. Analyse the resulting clusters. What do you conclude?
3. Choose a value of K and run the algorithm with different seeds. What is the
effect of changing the seed?
4. Run the EM algorithm on this data with the default parameters and describe the
output.
代写COSC2110/COSC2111 Data Mining
5. The EM algorithm can be quite sensitive to whether the data is normalized or
not. Usethewekanormalizefilter(Preprocess --> Filter --> unsupervised
--> normalize) to normalize the numeric attributes. What difference does
this make to the clustering runs?
6. The algorithm can be quite sensitive to the values of minLogLikelihoodImprove-
mentCV minStdDev and minLogLikelihoodImprovementIterating, Explore the effect
of changing these values. What do you conclude?
7. How many clusters do you think are in the data? Give an English language
description of one of them.
8. Compare the use Kmeans and EM for clustering tasks. Which do you think is
best? Why?
Data Mining 2 25-Jul-2016
9. What golden nuggets did you find, if any?
Submit: Up to one page that describes what you did for each of the above ques-
tions and your results and conclusions.
PART 4: ASSOCIATION FINDING 10 marks
1. The files supermarket1.arff and supermarket2.arff in the folder
/KDrive/SEH/SCSIT/Students/Courses/COSC2111/DataMining/data/arff
contain the same details of shopping transactions represented in two different
ways. You can use a text viewer to look at the files.
2. What is the difference in representations?
3. Load the file supermarket1.arff into weka and run the Apriori algorithm on
this data. You will need to restrict the number of attributes and/or the number
of examples. What significant associations can you find?
4. Exploredifferentpossibilitiesofthemetrictypeandassociatedparameters. What
do you find?
5. Load the file supermarket2.arff into weka and run the Apriori algorithm on
this data. What do you find?
6. Exploredifferentpossibilitiesofthemetrictypeandassociatedparameters. What
do you find?
7. Try the other associators. What are the differences to Apriori?
8. What golden nuggets did you find, if any?
9. [OPTIONAL] Can you find any meaningful associations in the bank data?
Submit: Up to one page that describes what you did for each of the above questions
and your results and conclusions.
Submission instructions: Submit through Blackboard assessment tasks.
Assessment Criteria: 70% of the marks are allocated for carrying out the runs and
reportingtheresults. 30%ofthemarksareforinvestigativestrategyandinterpretation
of the results.
Data Mining 3 25-Jul-2016
代写COSC2110/COSC2111 Data Mining
