Scribd
Upload
35 views

Business Analytics: Price Prediction of Mobile Phones Submitted To: Prof. Manit Mishra

This document presents the analysis and results of using a CART classification tree model to predict the price of mobile phones. Key findings include: - The confusion matrix shows the model correctly predicted 728 of 800 cases, with an accuracy of 91%. Sensitivity was 96.7% and specificity was 85.1%. - The fully grown tree used RAM, battery power, and pixel height to predict price. The best pruned tree used only RAM to predict price. - Lift charts show the model's predictions were an improvement over random guessing, with lift scores indicating which model performed best.

Uploaded by

Chayan Sen
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
35 views

Business Analytics: Price Prediction of Mobile Phones Submitted To: Prof. Manit Mishra

This document presents the analysis and results of using a CART classification tree model to predict the price of mobile phones. Key findings include: - The confusion matrix shows the model correctly predicted 728 of 800 cases, with an accuracy of 91%. Sensitivity was 96.7% and specificity was 85.1%. - The fully grown tree used RAM, battery power, and pixel height to predict price. The best pruned tree used only RAM to predict price. - Lift charts show the model's predictions were an improvement over random guessing, with lift scores indicating which model performed best.

Uploaded by

Chayan Sen
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

PGDM 2019-21

Business Analytics
Report
Price prediction of mobile phones

SUBMITTED TO: PROF. MANIT MISHRA

SUBMITTED BY:
Diksha lath 19pgdm-bhu023
Gourav kant Choudhury 19pgdm-bhu026
Kaustav kundu 19pgdm-bhu033
NItish ranjan 19pgdm-bhu045
Sauvik saha19pgdm-bhu060
Vaibhav gupta 19pgdm-bhu078
TABLE OF CONTENTS

Particulars Page
no.
 Analysis 2
 Confusion matrix
 Accuracy 3
 Sensitivity

 Specificity 4
 Precision
 F-value
 Fully Grown Tree 5
 Best Pruned Tree
 Lift Chart 6
 Decile-wise Lift Chart

1|Page
 ANALYSIS:

To predict the price of mobile phones, the predictive analysis used for the data
set is CART – Classification Tree as the dependent variable is categorical in
nature irrespective of the independent variables which are both continuous and
categorical in nature.

 CONFUSION MATRIX:

Confusion Matrix

  Predicted Class

Actual Class High Price Low Price


High Price 390 (TP) 13 (FN)

Low Price 59 (FP) 338 (TN)

Interpretation:
A confusion matrix here is used to describe the performance of a classification
model on the selected set of test data for which the true values are known.

2|Page
From the above confusion matrix,
we can say that out of 800 cases, there are 728 cases which are correctly
predicted and 72 cases which are wrongly predicted.
 390 cases which belong to success class and are correctly predicted also
called as true positives.
 338 cases which belong to failure class and are correctly predicted also
called as true negatives.
 13 cases belong to the actual class high price but are wrongly predicted to
be in success class also called as false negative.
 59 cases belong to the actual class low price but are wrongly predicted to
be in the failure class also called as false positive.

 Accuracy:

Accuracy (ACC) is calculated as the number of all correct predictions divided


by the total number of the dataset. The best accuracy is 1.0, whereas the worst is
0.0. It can also be calculated by 1 – ERR.

Accuracy = (390+338)/800 = 0.91

 Sensitivity (Recall or True positive rate):


Sensitivity (SN) is calculated as the number of correct positive predictions
divided by the total number of positives. It is also called recall (REC) or true
positive rate (TPR). The best sensitivity is 1.0, whereas the worst is 0.0.

= 390/ (390+13) = 0.967

3|Page
 Specificity (True negative rate):

Specificity (SP) is calculated as the number of correct negative predictions


divided by the total number of negatives. It is also called true negative rate
(TNR). The best specificity is 1.0, whereas the worst is 0.0.

= 338/ (338+59) = 0.851

 Precision (Positive predictive value):

Precision (PREC) is calculated as the number of correct positive predictions


divided by the total number of positive predictions. It is also called positive
predictive value (PPV). The best precision is 1.0, whereas the worst is 0.0.

= 390/ (390+59) = 0.868

 F-Value:

F-value is a harmonic mean of precision and recall. It usually varies between 0


and 1, the closer it is to 1 the better it is.

F-VALUE
= 0.91549

4|Page
 FULLY GROWN TREE:

If RAM >= 2261.5, then price of mobile is high.


If RAM >= 1481 and Battery Power >= 1433.5, then price of mobile is high.
If RAM >= 1481 and Battery Power < 1433.5 and Px Height >= 1131, then
price of mobile is high.
If RAM < 1481 and Px Height >= 1917, then price of mobile is high.

 BEST PRUNED TREE:

If RAM >= 2261.5, then price of mobile is high.

5|Page
 LIFT CHART and DECILE-WISE LIFT CHART:

A lift chart graphically represents the improvement that a mining model


provides when compared against a random guess and measures the change in
terms of a lift score. By comparing the lift scores for different models, we can
determine which model is best. We can also determine the point at which the
model’s predictions become less useful.

X-axis contains the cases and y-axis contains the true positives

Red line curve indicates the prediction without using the model and the blue
curve indicates the prediction with using the model.

6|Page

You might also like