ITC516: Data Mining and Visualization for Business Intelligence - Engineering Assignment Help

Assignment Task :

Task:

PART A: MULTIPLE CHOICE (30 marks)
Instructions:
• For each question, select the single best response from the options given. Circle the corresponding letter on your supplied GPAS.
• There is no negative marking for an incorrect answer. Attempt all questions. All questions have equal value (1 mark each).

1. Which of the following statements about the Naive Bayes algorithm is not correct?
A) Attributes are equally important
B) Attributes are statistically dependent of one another given the class value
C) Attributes are statistically independent of one another given the class value
D) Attributes can be nominal or numeric
2. What is meant by a cluster?
A) A group of similar objects that differ significantly from other objects
B) Symbolic representation of facts or ideas from which information can potentially be extracted
C) Operations on a database to transform or simplify data in order to prepare it for a machine-learning algorithm
D) A learning algorithm that is unsupervised
3. The task of inferring a model from labelled training data is called
A) Unsupervised learning
B) Supervised learning
C) Reinforcement learning
D) Deep learning
4. A telecommunication company wants to segment their customers into distinct groups in order to send appropriate subscription offers. This is an example of
A) Unsupervised learning
B) Supervised learning
C) Knowledge representation
D) Data transformation
5. Which of the following is not used in supervised learning algorithms?
A) Training data
B) Validation data
C) Knowledge data
D) Test data

6. What is meant by the term bias as used in data mining?
A) A class of learning algorithms that tries to find an optimum classification of a set of examples using probabilistic theory
B) A mechanism employed by a learning system to constrain the search space of a hypothesis
C) An approach to the design of learning algorithms that is inspired by human-based heuristics
D) A knowledge representation technique using tree structures
7. An optometrist wishes to use people’s characteristics to predict their contact lens type for a data mining system. In this case, the contact lens type would be an example of a
A) Class
B) Table
C) Feature
D) None of the above
8. The Euclidean measure is an example of a
A) Entropy function
B) Distance function
C) Bias function
D) Support function
9. What is meant by the dimensionality of data?
A) The size of the data
B) The structure of data in a file
C) The number of attributes or features
D) The levels of abstractions used to represent the data
10. What does the last attribute in an ARFF file refer to?
A) The file format
B) The class attribute of the data
C) The number of numeric attributes
D) An indication of whether the learning is supervised or unsupervised
11. What does a measure of purity refer to?
A) A measure for the choice on the next attribute to split on in a decision tree
B) The predictive power of a rule set
C) The impact of combining several models in data mining
D) The accuracy and completeness of the data
12. Which of the following is not found in a typical neural network structure?
A) Processing layer
B) Input layer
C) Output layer
D) Hidden layer

13. The diagram below shows an example of a

A) Classification matrix
B) Confusion matrix
C) Decision matrix
D) Correlation matrix
14. One way to reduce the “curse of dimensionality” for the k-Nearest Neighbor algorithm is to
A) Use a low value of k
B) Use a high value of k
C) Reduce the dimension of predictors
D) Use the Euclidean distance as the measure
15. Which of the following is a popular algorithm for training a neural network model?
A) Forward propagation
B) Forward learning
C) Back propagation
D) Back learning
16. Which of the following would be appropriate for a data mining algorithm aimed at discovering which groups of products consumers would tend to purchase together?
A) Decision trees
B) Association rules
C) Classification rules
D) 1R
17. What is meant by the term “curse of dimensionality”?
A) Affliction caused by adding samples to multi-variate data models
B) Affliction caused by adding variables to multi-variate data models
C) Advantage caused by adding samples to multi-variate data models
D) Advantage caused by adding variables to multi-variate data models
18. In a medical data mining system, a patient with the disease is classified as not having the disease. This is an example of a
A) True positive
B) True negative
C) False positive
D) False negative

19. An observation that is extreme, being distant from the rest of the data is termed a
A) Feature
B) Outlier
C) Predictor
D) Class
20. Which clustering algorithm starts with each cluster comprising exactly one observation and then progressively combines the two nearest clusters until there is just one cluster left at the end?
A) Agglomerative clustering
B) Divisive clustering
C) Single Linkage clustering
D) Complete Linkage clustering
21. The method of ensemble learning is to
A) Combine several models to improve performance
B) Combine several attributes to improve performance
C) Utilize advanced probabilistic models to improve performance
D) Utilize high performance computing systems to improve performance
22. What could be a definition of machine learning?
A) The autonomous acquisition of knowledge through the use of computer programs
B) The autonomous acquisition of knowledge through the use of manual programs
C) The selective acquisition of knowledge through the use of computer programs
D) The selective acquisition of knowledge through the use of manual programs

23. What would be the effect of averaging the output of multiple decision trees?
A) The bias would be increased
B) The bias would be decreased
C) The variance would be increased
D) The variance would be decreased
24. In a decision tree, which of the following is used to represent a segment?
A) Root node
B) Leaf node
C) Interior nodes
D) Exterior nodes

25. In a decision tree algorithm, which of the following is not correct for entropy?
A) A higher value of entropy denotes that the attribute values belong to mixed classes
B) The entropy value ranges from 0 to 1
C) An entropy value of 0 denotes that the attribute values are homogeneous
D) The objective is to select attributes that would increase the overall entropy value

26. The maximum margin classifier is associated with which of the following?
A) Linear regression
B) Logistic regression
C) Support vector machine
D) Decision tree
27. Which of the following represents a data mining procedure that tailors the model to work specifically for the given training data set only?
A) Overfitting
B) Underfitting
C) Generalization
D) Holdout
28. The goal of linear regression is to create a model that _______ the sum of squares of residuals or errors (SSE).
A) Averages
B) Logs
C) Minimizes
D) Maximizes
29. The logistic function is also termed as which of the following?
A) Gamma function
B) Sigmoid function
C) Epsilon
D) Sum of squares
30. In the k-Means algorithm, “k” refers to the
A) Number of instances
B) Number of attributes
C) Number of clusters
D) Number of dimensions

PART B: SHORT ANSWER (60 marks)
Instructions:
• Use the answer booklet to answer the questions in this part.
• Attempt any 10 questions. All questions have equal value (6 marks each).

1 With relation to data mining, identify whether the tasks below are supervised or unsupervised learning. Give a justification to your answer.
(a) Deciding whether to issue a loan to an applicant based on demographic and financial data (with reference to a database of similar data on prior customers). Supervised learning, because the database includes information on whether the loan was approved or not. (2 marks)

(b) In an online bookstore, making recommendations to customers concerning additional items to buy based on the buying patterns in prior transactions. Unsupervised learning, because there is no apparent outcome (e.g., whether the recommendation was adopted or not). (2 marks)

(c) Identifying a network data packet as dangerous (virus, hacker attack) based on comparison to other packets whose threat status is known. Supervised learning, because for the other packets the status is known. (2 marks)

2 Describe the difference in roles assumed by the training dataset, validation dataset and test dataset.
Training dataset is used to train or build models. For example, in a linear regression, the training dataset is used to fit the linear regression model, i.e. to compute the regression coefficients.

3 Consider the sample from a database of credit applicants in the table shown below. Comment on the likelihood that it was sampled randomly, and whether it is likely to be a useful sample.

Sample is not selected randomly as we can see from “observation #”, that there is pattern in the observations chosen for the sample. In particular, every 8th observation from the database was selected for the sample. (3 marks) When we select data with such a predecided methodology it might introduce a bias in the selected data set. This is true when the order of the observations in the dataset has some meaning (e.g., chronological order) (3 marks)

4 Describe the concept of overfitting. Explain why when a model is fit to training data, zero error with those data is not necessarily good.

5 A data mining routine has been applied to a transaction dataset and has classified 88 records as fraudulent (30 correctly so) and 952 as nonfraudulent (920 correctly so). Construct the classification confusion matrix and calculate the error rate.

(4 marks)

Error Rate = (32 + 58) / 1040 = 0.0865 = 8.65% (2 marks)
6 The table below shows a dataset containing 12 instances for a traffic injury system. Compute the naive Bayes conditional probability of an injury given WEATHER_R = 1 and TRAF_CON_R = 1, i.e. P(Injury=yes| WEATHER_R = 1, TRAF_CON_R =1.

7 List and describe two differences between classification trees and regression trees.
1. In classification trees the value of the leaf node is determined by the “voting? of the training data in that leaf. In regression trees the value of the leaf node is determined by the average of the training data. (3 marks)
2. Another difference is the measure of homogeneity used for choosing splits. (3 marks)

8 A neural net typically starts out with random coefficients. Hence, it produces essentially random predictions when presented with its first case. What is the key ingredient by which the net evolves to produce a more accurate prediction? Illustrate your answer with the backpropagation algorithm.

9 Describe a difference between association rules and classification rules. What are meant by the terms minimum support and confidence thresholds as used in the Apriori algorithm for generating association rules.

10 Discuss how missing values are to be handled in a data set, and how this impacts learning algorithms. Give an example when missing values in a dataset may be useful.
Description of handling missing values in a dataset (see the discussion beginning page 62 of the textbook) (4 marks)
Example when missing values may be useful – medical databases when missing values give indication of diagnosis. (2 marks)

11 Describe the k-means algorithm for clustering data. Is this a supervised or unsupervised learning algorithm?
Description of k-means algorithm (see the discussion beginning page 142 of the textbook) (4 marks)

k-means is an unsupervised learning algorithm. (2 marks)

12 When a bank that is in poor financial condition is misclassified as financially strong, the misclassification cost is much higher than when a financially strong bank is misclassified as weak. To minimize the expected cost of misclassification, should the cutoff value for classification (which is currently at 0.5) be increased or decreased? Justify your answer.

This ITC516 : Engineering Assignment has been solved by our Engineering Experts at My Uni Paper. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+ Students in Australia, UK & US by helping them to score HD in their academics. Our Experts are well trained to follow all marking rubrics & referencing style.

Be it a used or new solution, the quality of the work submitted by our assignment experts remains unhampered. You may continue to expect the same or even better quality with the used and new assignment solution files respectively. There’s one thing to be noticed that you could choose one between the two and acquire an HD either way. You could choose a new assignment solution file to get yourself an exclusive, plagiarism (with free Turnitin file), expert quality assignment or order an old solution file that was considered worthy of the highest distinction.