[go: up one dir, main page]
Scribd
Upload
76 views5 pages

Tutorial 8 Answers

The document describes using Naive Bayes classification to predict the likely outcome or class for new data points based on training data. It provides two examples - one predicting the class (approve loan or not) for applicants based on attributes like age, employment status, home ownership, and credit rating. The second predicts commute time (long, medium, short) between two locations based on hour of travel, weather, accidents, and vehicle stalls. For each example, it calculates the probabilities of each class given the new data point's attributes, and predicts the class with the highest probability based on independence assumptions of Naive Bayes.

Uploaded by

shimal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
76 views5 pages

Tutorial 8 Answers

The document describes using Naive Bayes classification to predict the likely outcome or class for new data points based on training data. It provides two examples - one predicting the class (approve loan or not) for applicants based on attributes like age, employment status, home ownership, and credit rating. The second predicts commute time (long, medium, short) between two locations based on hour of travel, weather, accidents, and vehicle stalls. For each example, it calculates the probabilities of each class given the new data point's attributes, and predicts the class with the highest probability based on independence assumptions of Naive Bayes.

Uploaded by

shimal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

IS328 Data Mining

Naïve Bayes Classification

Semester 2, 2019

Tutorial 8 Exercises

Q1. Consider the following loan database with 15 entries:

Using Naïve Bayes Classification, determine the likely outcome of the following two
applicants:
ID Age Has Own Credit Class
Job House Rating
16 middle True false good ??
17 young False true fair ??

SOLUTION
AGE HAS_JOB
YES NO YES NO
Young 2/9 3/6 True 5.9 0/6
Middle 3/9 2/6 False 4/9 6/6
Old 4/9 1/6

Own_House Credit_Rating
YES NO YES NO
True 6/9 0/6 Fair 1.9 4/6
False 3/9 6/6 Good 4/9 2/6
Excellent 4/9 0/6

P(Tes) = 9/15, P(No) = 6/15

E = age = middle , has_job = true, own_house = false, credit_rating = good


E1 is age = middle, E2 is has_job = true, E3 is own_house = false, E4 is credit_rating = good
We need to compute P(Yes } E), and P(No | E) and compare them.

P(Yes | E) =P(Yes), P(E1 | Yes) * P(E2 | Yes) * P(E3 | Yes)* P(E4 | Yes)
-------------------------------------------------------------------------------
P(E)

P(No | E) =P(No), P(E1 | No) * P(E2 | No) * P(E3 | No)* P(E4 | No)
-------------------------------------------------------------------------------
P(E)

P(E1 } Yes) = 3/9 P(E1 } No) = 2/6


P(E2 } Yes) = 5/9 P(E2 } No) = 0/6
(5+1) / (9+2) = 6/11 (0+1)/(6+2) = 1/8
P(E3 } Yes) = 3/9 P(E3 } No) = 6/6
P(E4 } Yes) = 4/9 P(E4 | No) = 2/6
P(Yes)= 9/15 P(No) = 6/15

P( Yes | E) = 9/15 * 3/9 * 6/11 * 3/9 * 4/9 = 0.0162


P(No | E) = 6/15 * 2/6 * 1/8 * 6/6 * 2/6 = 0.0056
Hence, the Naïve Bayes Classifier predicts the outcome as Yes for the example.
Q2 Given the training data given below (Travel time between the USP and the Nausori
Airport during morning peak time ) predict the class of the following example using Naïve
Bayes Classification.

hour:10 am , weather = rainy, accident = no, stall = no

Exampl
e Attributes Target

  Hour Weather Accident Stall Commute

D1 8 AM Sunny No No Long

D2 8 AM Cloudy No Yes Long

D3 10 AM Sunny No No Short

D4 9 AM Rainy Yes No Long

D5 9 AM Sunny Yes Yes Long

D6 10 AM Sunny No Yes Medium

D7 10 AM Cloudy No No Short

D8 9 AM Rainy No No Medium

D9 9 AM Sunny Yes No Long

D10 10 AM Cloudy Yes Yes Long

D11 10 AM Rainy No No Short

D12 8 AM Cloudy Yes No Long

D13 9 AM Sunny No No Medium

SOLUTION
E = hour:10 am , weather = rainy, accident = no, stall = no
E1 is hour = 10 am, E2 is weather = rainy, E3 is accident = no, E4 is stall = no
We need to compute P(long } E), P (medium | E) and P(Short | E) and compare them.
P(l ong | E) =P(long), P(E1 | long) * P(E2 | long) * P(E3 | long)* P(E4 | long)
-------------------------------------------------------------------------------
P(E)

P(E1 } long) = 1/7 P(E1 } medium) = 1/3 P(E1 | short) = 3/3

P(E2 } long) = 1/7 P(E2 } medium) = 1/3 P(E2 | short) = 1/3

P(E3 } long) = 2/7 P(E3 } medium) = 2/3 P(E3 | short) = 3/3

P(E4 } long) = 4/7 P(E4 | medium) = 2/3 P(E4 | short) = 3/3

P( Long | E) = 7/13 * 1/7 * 1/7 * 2/7 * 4/7 = 8 /4459= 0.0018


P(Medium | E) = 3/13 * 1/3 * 1/3 * 2/3 * 2/3 = 4/351= 0.0114
P(Short | E) = 3/13 * 3/3 * 1/3 * 3/3 * 3/3 = 1/13 = 0.0767
Hence, thw Naïve Bayes Classifier predicts the commute time as short for the example.

You might also like