Week 15 Assignment
Week 15 Assignment
In the first part, you will get familiar with explainability and interpretability methods
used to understand ML models decisions. You can make use of the boiler plate code
to begin with.
In part-II, you will learn to develop signature verification using the RSA Algorithm. You
can make use of the boiler plate code to begin with.
Part-I
Introduction:
In this assignment, we will aim to model bike rental prediction using Machine Learning
and explore the explainability of the model.
Dataset:
http://archive.ics.uci.edu/ml/datasets/Bike+Sharing+Dataset
Task:
explainable_AI_starter.ipynb
1. Understand how the h2o package helps in explainability with various plots on
relation between attributes and the defect prediction. Comment on your findings.
2. Try modifying the code on Bike rental dataset and see how the package tries to
explain the amount of bikes rented correspond to the various environmental
conditions.
Part-II
Introduction:
In this assignment, we will aim to develop a signature verification protocol using the
RSA algorithm.
The RSA public-key cryptosystem provides a digital signature scheme (sign + verify),
based on the math of the modular exponentiations and discrete logarithms and the
computational difficulty of the RSA problem.
● Key Generation:- The RSA key-pair consists of: public key {n, e} & private key
{n, d}. The numbers n and d are typically big integers, while e is small. By
definition, the RSA key-pairs has the following property:
● RSA Sign:- sign a message ‘msg’ with the private key components {n,d}
○ Calculate the message hash: h = hash(msg)
○ Encrypt h to calculate the signature: s = h^d (mod n)
● RSA Verify Signature:- Verify a signature s for the message ‘msg’ with the
public key {n, e}
○ Calculate the message hash: h = hash(msg)
○ Decrypt the signature: h′ =s^e (mod n)
○ Compare h with h' to find whether the signature is valid or not
Task:
Signature_Verification_Starter.ipynb
Note:- Try to save all the keys, messages, and signature in the files.
Note: The file should be submitted in .ipynb and .HTML (both) formats.
Regards,
Program Office