Download Test

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Cluster analysis wikipedia , lookup

Nonlinear dimensionality reduction wikipedia , lookup

Transcript
IT 241
Information Architecture Exam 3
Page 1
December 2, 2009
Name _____________________________
This exam is open book and on-line but no contact with a live person. Be careful with your time!
1. In the Juniata.edu website distinguish between the global and local navigation systems.
[5 pts]
2. Describe how you would stress test a website regarding navigation?
[5 pts]
2. Assume we have a set of documents in our web site and a search results in a 20% recall ratio and an 80%
precision ratio.
[10 pts]
a.) If there were 100 relevant documents possible, how many relevant documents were actually retrieved in
the search?
_______________
b.) How many documents were retrieved? ___________________
c.) Explain why it is difficult to have both a high recall ratio and precision ratio.
IT 241
Information Architecture Exam 3
Page 2
3. What is a continuous interaction type application and give an example.
[5 pts]
4. Distinguish between stepped and manual interaction modes by giving an example of each from the Weather
Channel site www.weather.com.
[5 pts]
5. Distinguish between exploratory and involuntary intentions of interaction by giving an example of from the
Arch.
[5 pts]
6. Why are residue and scent something to strive for in a web site?
[5 pts]
7. How can clustering be used as an early step in the data mining and discovery process? What do clusters tell
you?
[5 pts]
IT 241
Information Architecture Exam 3
Page 3
[7 pts]
8. Given the decision tree rule for the above dataset
IF Sex=Female && IncomeRange=30-40K
THEN CreditCardInsurance=Yes
Determine its accuracy = ___________% and its coverage = ____________ %
Along with these two rules
IF Sex=Male
THEN CreditCardInsurance=No
IF Sex=Female && IncomeRange != 30-40K
THEN CreditCardInsurance=No
Draw a decision tree to correspond
with these three production rules.
9. Given the confusion matrix.
[8 pts]
a.
b.
c.
d.
Cat
Dog
Rabbit
Cat
12
6
0
Dog
4
6
2
Rabbit
0
4
16
Number of cats total= ____________
Number of cats classified as a dog = _______
Number of dogs incorrectly classified = ________
Percent classified correctly (all three categories) = ________
IT 241
Information Architecture Exam 3
Page 4
10. When determining the attributes to be organized into a decision tree, describe the process and criterion used to
choose the first attribute and then those that follow on the next level.
[5 pts]
11. Explain what or why one might do these preprocessing activities to prepare a raw dataset for datamining.
[15 pts]
a. Noisy data correction--
b. Dealing with missing data --
c. Normalization (not database) --
d. Data type conversion --
e. Attribute creation --
12. True/false.
[10 pts]
_____ A browser’s navigation features should be leveraged for most web sites.
_____ Icons are discouraged in navigation systems unless complemented with textual labels.
_____ Searchable web page content should be limited to the metatext only and ignore tag contents.
_____ Research shows that navigating a menu system with limited options is more accurate than a deeper one
with many options per level.
_____.Visual transitions in interactive displays help combat the change and inattentional blindness phenomena.
_____.A data cube for data mining is created by multiple joins of tables from the operational and/or archival
database.
_____ A datacube should not contain repetitive data.
_____.Linear regression requires all attributes to be numerical.
IT 241
Information Architecture Exam 3
Page 5
_____.Linear regression modeling can be used as an attribute selection process.
_____.Norman’s Action Cycle is useful to evaluate interaction models.
_____ Brute force algorithms for attribute selection are limited to small sets of attributes because of exponential
growth.
13. Briefly describe what results from applying each of these operations (separately) on the pivot table or data
cube below.
[10 pts]
a. Rollup on the region
b. Drill down on Travel-Q4-Region One
c. Slice on Retail