0% found this document useful (0 votes)
74 views

Assignment Projects-13th Nov

The document provides instructions for submitting assignments and projects by November 13th, 2020. It describes 4 projects involving predicting company profits, wine classification, disease prediction, and iris flower clustering. It also includes an exercise on K-Nearest Neighbors algorithm and implementation using diabetes data.

Uploaded by

venkiscribd444
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
74 views

Assignment Projects-13th Nov

The document provides instructions for submitting assignments and projects by November 13th, 2020. It describes 4 projects involving predicting company profits, wine classification, disease prediction, and iris flower clustering. It also includes an exercise on K-Nearest Neighbors algorithm and implementation using diabetes data.

Uploaded by

venkiscribd444
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Assignment & Projects

Instructions

• Please submit the assignment & projects before 13th November, 2020, 06:00
PM. It should be submitted to LMS only, submission through other means
will not be accepted.
• Any submission after that will not be accepted.
• All assignment & projects should be submitted in a single file (No separate
file for each assignment & projects)
• File should be in .ipynb format only. Any other format will not be accepted.
• Copy and Paste in any form is not recommended and will have negative
impact on the evaluation of project.
• New idea and concept will be highly appreciated.
• Practice: Exercise 1 is for practice and is not Considered for Evaluation.
Assignment 1:

• Consider a case of 50 student, out of which 35Passed (P) and 15 Failed


(F) in the Certification Exam. Write a program to compute Entropy for
the given information.

• If we have the information that some of the candidates have undergone


training and some didn’t. Based on it split the dataset, let say 20 people
undergone the training and 30 didn’t. Out of this 20 trained candidate
18P and 2F, while in case of 30 untrained 17P and 13F. Write a
program to calculate the information gain after split of data.
Project 1:
From Data given it is required to develop the model to predict the profit of the company. Dataset file
name: Companies-profit.csv

Task:
1) Load the dataset
2) Analyse the data visually and give justification for selecting particular algorithm for developing the
model.
3) Split it into test and train.
4) Develop a model for predicting the profit of the company
5) Measure all the parameters of the model
6) Measure the performance of the model
Project 2:
• Use wine dataset from sklearn
• Load the dataset and split it into test and train.
• Develop the model to Classify wines into different categories.
• Train the model using Gaussian and Multinominal classifier, please mention
which model is better and why.
• Use the trained model to perform some predictions on test data.
Project 3:
• Datasets for Disease operation is provided. Dataset file name: Disease.csv
1. Load the dataset and split it into test and train.
2. Analyse the Data Visually.
3. Find limitation with data if any in developing the model
4. Develop the model to identify whether operation successful or not using
Decision Tree algorithm
5. Measure the performance of the model
6. Try to improvise the model by suitable technique.
7. Measure the performance of the model
Project: 4
• Use iris flower dataset from
sklearn library and try to
form clusters of flowers
using petal width and length
features.
• Analyse the Data Visually
before training
• Drop other two features for
simplicity.
• Try to Use K Mean
Clustering
• Analyse the Data Visually by
comparing before and after
the training.
Practice: Exercise 1
1. What is K-Nearest Neighbour(KNN) Algorithm for Machine Learning
2. How does K-NN work?
3. Implementation of the KNN algorithm- Take dataset: diabetes.csv
• Data Pre-processing step-if any
• Fitting the K-NN algorithm to the Training set
• Predicting the test result
• Test accuracy of the result

You might also like