The document provides 9 exercises related to machine learning algorithms including backpropagation neural networks, Naive Bayes, decision trees, k-NN, and associative classification. The exercises involve designing neural networks to predict credit ratings, analyzing sample neural network architectures, explaining why Naive Bayes is called "naive", performing classification using Naive Bayes on sample datasets, comparing Naive Bayes and neural networks using Weka, and applying k-NN, decision trees and associative classification on sample datasets.
The document provides 9 exercises related to machine learning algorithms including backpropagation neural networks, Naive Bayes, decision trees, k-NN, and associative classification. The exercises involve designing neural networks to predict credit ratings, analyzing sample neural network architectures, explaining why Naive Bayes is called "naive", performing classification using Naive Bayes on sample datasets, comparing Naive Bayes and neural networks using Weka, and applying k-NN, decision trees and associative classification on sample datasets.
The document provides 9 exercises related to machine learning algorithms including backpropagation neural networks, Naive Bayes, decision trees, k-NN, and associative classification. The exercises involve designing neural networks to predict credit ratings, analyzing sample neural network architectures, explaining why Naive Bayes is called "naive", performing classification using Naive Bayes on sample datasets, comparing Naive Bayes and neural networks using Weka, and applying k-NN, decision trees and associative classification on sample datasets.
The document provides 9 exercises related to machine learning algorithms including backpropagation neural networks, Naive Bayes, decision trees, k-NN, and associative classification. The exercises involve designing neural networks to predict credit ratings, analyzing sample neural network architectures, explaining why Naive Bayes is called "naive", performing classification using Naive Bayes on sample datasets, comparing Naive Bayes and neural networks using Weka, and applying k-NN, decision trees and associative classification on sample datasets.
Classification. Exercise 1. Suppose we want to classify potential bank customers as good creditors or bad creditors for loan applications. We have a training dataset describing past customers using the following attributes: Marital status {married, single, divorced}, Gender {male, female}, Age {[18..30[, [30..50[, [50..65[, [65+]}, Income {[10K..25K[, [25K..50K[, [50K..65K[, [65K..100K[, [100K+]}. Design a neural network that could be trained to predict the credit rating of an applicant. Exercise 2. Given the following neural network with initialized weights as in the picture, explain the network architecture knowing that we are trying to distinguish between nails and screws and an example of training tupples is as follows: T1{0.6, 0.1, nail}, T2{0.2, 0.3, screw}.
6 -0.4
3 0.1
0.2
-0.1
-0.2 0.1
0.1 0
0.2
0.5
0.2 0.3
hidden neurons
0.6
-0.1 4
-0.2
output neurons
0.6
-0.4
input neurons
Let the learning rate be 0.1 and the weights be as indicated in the figure above. Do the forward propagation of the signals in the network using T1 as input, then perform the back propagation of the error. Show the changes of the weights.
Exercise 3. Why is the Nave Bayesian classification called nave?
Exercise 4. Nave Bayes for data with nominal attributes
Given the training data in the table below (Buy Computer data), predict the class of the following new example using Nave Bayes classification: age<=30, income=medium, student=yes, credit-rating=fair
Exercise 5. Applying Nave Bayes to data with numerical attributes and using the Laplace correction (to be done at your own time, not in class) Given the training data in the table below (Tennis data with some numerical attributes), predict the class of the following new example using Nave Bayes classification: outlook=overcast, temperature=60, humidity=62, windy=false. Tip. You can use Excel or Matlab for the calculations of logarithm, mean and standard deviation. The following Matlab functions can be used: log2 logarithm with base 2, mean mean value, std standard deviation. Type help <function name> (e.g. help mean) for help on how to use the functions and examples.
Exercise 6. Using Weka (to be done at your own time, not in class) Load iris data (iris.arff). Choose 10-fold cross validation. Run the Nave Bayes and Multi-layer percepton (trained with the backpropagation algorithm) classifiers and compare their performance. Which classifier produced the most accurate classification? Which one learns faster?
Exercise 7. k-Nearest neighbours
Given the training data in Exercise 4 (Buy Computer data), predict the class of the following new example using k-Nearest Neighbour for k=5: age<=30, income=medium, student=yes, creditrating=fair. For distance measure between neighbours use a simple match of attribute values: distance(A,B)=
w * (a , b ) i =1
4 where (ai , bi ) is 1 if ai equals bi and 0 otherwise. ai and bi are
either age, income, student or credit_rating. Weights are all 1 except for income it is 2.
Exercise 8. Decision trees
Given the training data in Exercise 4 (Buy Computer data), build a decision tree and predict the class of the following new example: age<=30, income=medium, student=yes, credit-rating=fair.
Exercise 9. Associative Classifier
Given the training data in Exercise 4 (Buy Computer data), build an associative classifier model by generating all relevant association rules with support and confidence thresholds 10% and 60% respectively. Classify using this model the new example: age<=30, income=medium, student=yes, credit-rating=fair, selecting the rule with the highest confidence. What would be the classification if we chose to vote the class among all rules that apply?
