Skip to content

C++ implementations of various classification algorithms on the "banknote authentication Data Set".

License

Notifications You must be signed in to change notification settings

shubhamjha97/ml2-assignment

Repository files navigation

Selected Topics in Computer Science

Course Assignment for CS F441- Selected Topics from Computer Science @ BITS Pilani, Hyderabad Campus.

Done under the guidance of Dr. N.L. Bhanu Murthy, Associate Professor, BITS Pilani, Hyderabad Campus.

Assignment 1

C++ implementations of various classification algorithms on the "banknote authentication Data Set". The dataset can be found here.

1) Fisher's Linear Discriminant Analysis

Threshold: 0.664737

Precision : 1

Recall : 0.972973

Accuracy : 98.7864 %

Confusion matrix

n=412 Predicted: POSITIVE Predicted: NEGATIVE
Actual: POSITIVE 180 5
Actual: NEGATIVE 0 227

2) Probabilistic Generative Model

Precision : 0.866279

Recall : 0.805405

Accuracy : 85.6796 %

Confusion matrix

n=412 Predicted: POSITIVE Predicted: NEGATIVE
Actual: POSITIVE 149 36
Actual: NEGATIVE 23 204

3) Logistic Regression

Precision : 0.984043

Recall : 1

Accuracy : 99.2718 %

Confusion matrix

n=412 Predicted: POSITIVE Predicted: NEGATIVE
Actual: POSITIVE 185 0
Actual: NEGATIVE 3 224

Group Members:

Shubham Jha (2015A3PS0288H)

Saurabh Khandelwal (2015A7PS0082H)

Aditya Desai (2015A3PS0211H)

About

C++ implementations of various classification algorithms on the "banknote authentication Data Set".

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages