Easy

Question 1

Distance Functions, Euclidean, _, _

Answer 1

Manhattan, Minkowski

Question 2

optimal dataset for K neighbours

Answer 2

3-10

Question 3

association rules find all sets of itemsets that have _ greater than the minimum

Answer 3

support

Question 4

association rules: find desired rules that have _ greater than the min

Answer 4

confidence

Question 5

association rules are usually needed to satisfy a user-specified _ & _

Answer 5

minimum support, confidence

Question 6

formula: support for association rules

Answer 6

frq(x,y)/n

Question 7

formula: confidence for association rules

Answer 7

frq(x,y)/frq(x)

Question 8

K-means clustering… place _ at _ locations; repeat until convergence

Answer 8

centroids, random

Question 9

K-means clustering… 1. for each point xi:

Answer 9

find nearest centroid

Question 10

K-means clustering… 2. assign the point, & for each determine new centroid

Answer 10

to cluster

Question 11

K-means clustering… 3. stop when non of the __ change

Answer 11

clustering assignments

Question 12

a perceptron is used to classify _ classes

Answer 12

linearly separable

Question 13

a percepton consists of _, _, _

Answer 13

weights, summation processor, activation function

Question 14

a percepton takes a weighted sum of input and outputs, 1 if the sum > than _ _ _ _, _

Answer 14

some adjusted threshold value, theta

Question 15

the perceptron can have another input known as

Answer 15

the bias

Question 16

perceptron: it is normal practice to treat the bias as

Answer 16

just another input

Question 17

the perceptron bias allows us to

Answer 17

shift the transfer curve horizontally along the input axis

Question 18

the perceptron weights determine

Answer 18

the slope of the curve

Question 19

draw the perceptron

Answer 19

Question 20

perceptron concept: the ouput is set at one of two levels, depending on whether the _, is greater or less than some _ value. This is called:

Answer 20

total input, threshold, unit step (threshold)

Question 21

draw the unit step threshold

Answer 21

Question 22

Perceptron function: the _ consists of two functions, _ and _ , ranging from 0 and 1, and -1 to +1

Answer 22

sigmoid, logistic, tangential

Question 23

perceptron function: Ouput is proportional to the total weighted output

Answer 23

piecewise linear

Question 24

perceptron function: bell shaped curves that are continuos. the node output (high / low) is interpreted in terms of class membership (1/0) depending on how close the net input is to a _

Answer 24

Gaussian, chosen value of average

Question 25

what helps us control how much we change the weight and bias in a perceptron, which we do in order to get a smallest error

Answer 25

the learning rate

Question 26

Perceptron: if we have n variables then we need to find _

Answer 26

n + 1 weight values (n variables + the bias)

Question 27

perceptron: if we have 2 inputs the equation becomes:

Answer 27

w1x1 + w2x2 + b = 0

where wi is the weight of input i and b is the bias (w0 with input value x0 of 1)

Question 28

what is the data mining task of prediction the value of a categorical variable (target or class)

Answer 28

classification

Question 29

transforming attributes from numerical to categorical

Answer 29

binning or discretization

Question 30

transforming attributes from categorical to numerical

Answer 30

encoding or continuization

Question 31

Represents

Answer 31

Linear / Non-linear separability & inseperable

Question 32

list the frequency table methods

Answer 32

• ZeroR
• One R
• Naive Bayesian
• Decision Tree

Question 33

list the covariance matrix methods of classification

Answer 33

• linear discriminate analysis
• logistic regression

Question 34

list the similarity functions method of classification

Answer 34

K Nearest Neighbours

Question 35

List the other methods of classication

Answer 35

artificial neural network, support vector machines

Question 36

simplest classification method

Answer 36

ZeroR

Question 37

Zero R classifier relies:

Answer 37

on the targer and ignores all predictors

Question 38

although there is no predictability power in ZeroR it is useful for _

Answer 38

determining a baseline performance as a benchmark for other classification methods

Question 39

how to implement ZeroR

Answer 39

contruct a frequency table for the target and select it’s most frequent value

Question 40

Zero R only predicts _

Answer 40

the majority class correctly (as shown by the confusion table)