Consider the hypothesis space defined over instances shown below, we characterize eac hypothesis (apple taste) by 4-tuples. Please hand trace the ID3 classifier to build a decisio tree, then predict the target value Taste=Sweet/Tart for the following instances: a) b) c) d) e) Now suppose the actual taste of the five apples above are actually "Sweet, Sweet, Sweet, Tart, Tart", what is the accuracy of the decision tree? Please show all the steps and include the corresponding confusion matrix for accuracy calculation. (10pts) Melon Color 1 Red 2 Red 3 Yellow 4 Yellow 5 Yellow 6 Yellow 7 Green 8 Red 9 Red 10 Yellow Crispiness Spot Fragrant None Yes None No None Yes Some No None No Some Yes Some No None Yes None No None Yes High High High Mid Low Mid Low Low Low Mid Taste Sweet Sweet Sweet Tart Sweet Tart Sweet Tart Sweet Sweet

Consider the hypothesis space defined over instances shown below, we characterize eac hypothesis (apple taste) by 4-tuples. Please hand trace the ID3 classifier to build a decisio tree, then predict the target value Taste=Sweet/Tart for the following instances: a) b) c) d) e) Now suppose the actual taste of the five apples above are actually "Sweet, Sweet, Sweet, Tart, Tart", what is the accuracy of the decision tree? Please show all the steps and include the corresponding confusion matrix for accuracy calculation. (10pts) Melon Color 1 Red 2 Red 3 Yellow 4 Yellow 5 Yellow 6 Yellow 7 Green 8 Red 9 Red 10 Yellow Crispiness Spot Fragrant None Yes None No None Yes Some No None No Some Yes Some No None Yes None No None Yes High High High Mid Low Mid Low Low Low Mid Taste Sweet Sweet Sweet Tart Sweet Tart Sweet Tart Sweet Sweet

Database System Concepts

7th Edition

ISBN:9780078022159

Author:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Publisher:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Chapter1: Introduction

Section: Chapter Questions

Problem 1PE

See similar textbooks

Similar questions

Load the soybean diagnosis data set in Weka (found in Weka-3.6/data/soybean.arff), then perform the following: Build a decision tree by selecting J48 as the classifier and 10-way cross-validation. Then fill out the following table: Correctly Classified Instances Incorrectly Classified Instances Kappa statistic Mean absolute error Root mean squared error Relative absolute error Root relative squared error Total Number of Instances Build a Naïve Bayes classifier and select 10-way cross-validation. Then fill out the following table: Correctly Classified Instances Incorrectly Classified Instances Kappa statistic Mean absolute error Root mean squared error Relative absolute error Root relative squared error Total Number of Instances Compare between results in previous two sections (a and b), which algorithm give the better result and…
Sketch a Voronoi map for the INN method for the following examples, and indicate the decision surface that results. What would the decision surface look like if you used the prototype based approach (replace the set of positive examples by a single example lying in the middle of them, same for the negatives)? Think about some other differences between the prototype based approach and storing each individual example. + + + + + + +
Question 47.Random forests are one of the most famous machine learning methods. They are easy to understand,easy to implement and reach good prediction performances even without a hyper-parameter tuning. Which of thefollowing statements on random forest are correct?a) The prediction of a classification forest is made by a majority vote of the trees’ predictions.b) The prediction of a regression forest is the median of the tree predictions.c) Each single tree in the forest uses only a part of the data available.d) The training time of a random forest scales linear with the number of trees used.
extra credit only for section 412 , required for other sections) Prove that if a heuristic is consistent, it must be admissible. Construct an admissible heuristic that is not consistent. Hint: For the first part, use induction on the number of steps between a nodenand the goal .
Implement the FIND-S algorithm. Now use this program to study the number of random training examples required to exactly learn the target concept. Implement a training example generator that generates random instances, then classifies them according to the target concept: (Sunny, Warm?, ?, ?, ?) Consider training your FIND-Sprogram on randomly generated examples and measuring the number of examples required before the program's hypothesis is identical to the target concept. Can you predict the average number of examples required? Run the experiment at least 20 times and report the mean number of examples re- quired. How do you expect this number to vary with the number of "?" in the target concept? How would it vary with the number of attributes used to describe instances and hypotheses?
3. For a given data set, we applied the 0.632 bootstrapping validation and built a decision tree model. The training error was computed as 0.02 and the testing set error is equal to 0.3. What is the overall error estimate for the model?
Question 47. Random forests are one of the most famous machine learning methods. They are easy to understand, easy to implement and reach good prediction performances even without a hyper-parameter tuning. Which of the following statements on random forest are correct? a) The prediction of a classification forest is made by a majority vote of the trees' predictions. b) The prediction of a regression forest is the median of the tree predictions. c) Each single tree in the forest uses only a part of the data available. d) The training time of a random forest scales linear with the number of trees used.
. Imagine you are given the training instances one at a time, instead of all at once. How can you incrementally adjust S and G in such a case? (Hint: See the candidate elimination algorithm in Mitchell 1997.)
Problem 6 Consider a data set that contains 4 Boolean attributes A, B, C, and D, and a Boolean class y. For the Boolean expression below (between the class the rest of the attributes), it is possible to construct a smaller decision tree that perfectly classifies the data without generating the complete tree (i.e., the number of leave nodes is less than 16). Then, draw such a tree. and y = A ^ BA C ^ D
Given the dataset of cars with their mileage and cost for cost prediction: Build a decision tree and attach the picture of you decision tree formed as the solution to the question using hints from tutorial_6. What do you think is special about the data type of the values being predicted in this problem? Do a search and find out why a DecisionTreeRegressor is used in this case (instead of the general purpose DecisionTreeClassifier)? Both examples are in the tutorial code. array([['Toyota Corolla', '40', '20175'], ['Ford', '45', '25000'], ['Dodge', '62', '35782'], ['Chevrolet', '50', '30000'], ['Canoo', '57', '34750'], ['Tesla', '113', '54000'], ['BMW', '70', '36400']], dtype='<U21')
2 You are given the count matrix in the table 4 for the attribute "Previous GPA". The class label is whether a student will pass (C1) or fail (CO) a class. We want to use the attribute "Previous GPA" asour splitting attribute in an inner node of a decision tree with a binary test condition. Which is the best way to partition the attribute values into two groups? Form the count matrices for the different partitions and compute the classification error. Explain your reasoning while considering the metric of classification error to evaluate the different partitions. 0 3.1 + 0 3 Consider the decision trees shown in Figure 1. The decision tree in 1b is a pruned version of the original decision tree 1a. The training and test sets are shown in table 5. For every combination of values for attributes A and B, we have the number of instances in our dataset that have a positive or negative label. B 2 2 1 0 1 0 1 0 1 0 + 5 A 3 22 7 1 2 6 Bad 1 pass (C1) fail (CO) 4 B (a) Decision Tree 1 (DT1) 1 3…
Consider the hypothesis space defined over instances shown below, we characterize each hypothesis (apple taste) by 4-tuples. Please hand trace the ID3 classifier to build a decision tree, then predict the target value Taste=Sweet/Tart for the following instances: a) <Red, High, Some, No> b) <Red, Low, Some, Yes> c) <Yellow, Low, Some, No> d) <Green, High, None, No> e) <Green, Mid, Some, Yes> Now suppose the actual taste of the five apples above are actually “Sweet, Sweet, Sweet, Tart, Tart”, what is the accuracy of the decision tree? Please show all the steps and include the corresponding confusion matrix for accuracy calculation. (please................ as soon as give correct solution)