Scribd
Upload
61 views2 pages

COMP 428 Machine Learning Assignments

The document outlines a series of tasks related to machine learning, including programming exercises for data handling, sentiment analysis, and neural network training. It covers practical applications of algorithms, parameter calculations for neural networks, and recommendations for classification methods based on feature complexity. Additionally, it includes a recruitment agency's challenge of predicting salaries based on experience, with specific programming tasks to analyze and visualize data.

Uploaded by

sitemba4
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
61 views2 pages

COMP 428 Machine Learning Assignments

The document outlines a series of tasks related to machine learning, including programming exercises for data handling, sentiment analysis, and neural network training. It covers practical applications of algorithms, parameter calculations for neural networks, and recommendations for classification methods based on feature complexity. Additionally, it includes a recruitment agency's challenge of predicting salaries based on experience, with specific programming tasks to analyze and visualize data.

Uploaded by

sitemba4
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

COMP 428 MACHINE LEARNING CAT ONE

a) Suppose you are given the following data set in CSV format

i) Write a python program to load the data set make assumption where necessary (3Marks)

ii) Write a program to extract dependent and independent variable (2Marks)

iii) Write a program to handle missing data (2Marks)

b) One of the application areas of naïve bayes algorithm is sentiment analysis, using an example
explain the meaning of the term sentimental analysis. (4Marks)

c) Discuss the types of the machine learning techniques (6Marks)

a) You want to train a neural network to drive a car. Your training data consists of grayscale 64 ×64
pixel images. The training labels include the human driver’s steering wheel angle in degrees and
the human driver’s speed in miles per hour. Your neural network consists of an input layer with 64
× 64 = 4,096 units, a hidden layer with 2,048 units, and an output layer with 2 units (one for steering
angle, one for speed). You use the ReLU activation function for the hidden units and no activation
function for the outputs (or inputs).
Calculate the number of parameters (weights) in this network. (6Marks)
d) A 4 - input neuron has weight 1 2 3 4 , The transfer function is linear with a constant of
proportionality of 2 , the inputs are , 4, 10 , 5 and 20, Calculate the output. (3Marks)

e) Some patient features are expensive to collect (e.g., brain scans, heart etc ) whereas others are
not (e.g., temperature, age ,Bp) and want our classification algorithm to predict whether a patient
has a given disease based on the less complicated features, if the classifier is 80% confident then
the patient has that disease, afterwards we can do additional examinations to collect additional
patient features to improve the accuracy. In this case, which classification methods do you
recommend for such a task , neural networks, decision tree, or naive Bayes? Justify your

choice. (4Marks)

f) A recruitment agency faces a challenge of determining the salary of an employee after several
years of experience, write a program in python that will help the agency achieve the following
tasks based on a data set
i) Import the necessary libraries (2Marks)
ii) Load data training set (2Marks)
iii) Find out how many rows and columns are in the dataset (2Marks)
iv) Find out the statistical summary (2Marks)
v) Visualize the data set (2Marks)

You might also like