Scribd
Upload
426 views3 pages

CSC583 Artificial Intelligence Algorithms Group Assignment (30%)

Uploaded by

harith danish

This document outlines a group assignment to classify bank notes as genuine or forged using a neural network model. Students are provided a dataset containing features extracted from bank note images and asked to: 1) explore and prepare the data, 2) split it into train and test sets, 3) build a neural network classifier, 4) generate a confusion matrix to evaluate performance, and 5) submit a single PDF with their work. The assignment aims to accurately detect fraudulent bank notes and evaluate a neural network's ability for classification problems.

Copyright:

© All Rights Reserved
Download as PDF, TXT or read online from Scribd
Download as pdf or txt

CSC583 Artificial Intelligence Algorithms Group Assignment (30%)

Uploaded by

harith danish
426 views3 pages
This document outlines a group assignment to classify bank notes as genuine or forged using a neural network model. Students are provided a dataset containing features extracted from bank note images and asked to: 1) explore and prepare the data, 2) split it into train and test sets, 3) build a neural network classifier, 4) generate a confusion matrix to evaluate performance, and 5) submit a single PDF with their work. The assignment aims to accurately detect fraudulent bank notes and evaluate a neural network's ability for classification problems.

Original Title

GROUPASSIGNMENT

Copyright

© © All Rights Reserved

Available Formats

PDF, TXT or read online from Scribd

Did you find this document useful?

Is this content inappropriate?

This document outlines a group assignment to classify bank notes as genuine or forged using a neural network model. Students are provided a dataset containing features extracted from bank note images and asked to: 1) explore and prepare the data, 2) split it into train and test sets, 3) build a neural network classifier, 4) generate a confusion matrix to evaluate performance, and 5) submit a single PDF with their work. The assignment aims to accurately detect fraudulent bank notes and evaluate a neural network's ability for classification problems.

Copyright:

© All Rights Reserved
Download as PDF, TXT or read online from Scribd
Download as pdf or txt
426 views3 pages

CSC583 Artificial Intelligence Algorithms Group Assignment (30%)

Uploaded by

harith danish
This document outlines a group assignment to classify bank notes as genuine or forged using a neural network model. Students are provided a dataset containing features extracted from bank note images and asked to: 1) explore and prepare the data, 2) split it into train and test sets, 3) build a neural network classifier, 4) generate a confusion matrix to evaluate performance, and 5) submit a single PDF with their work. The assignment aims to accurately detect fraudulent bank notes and evaluate a neural network's ability for classification problems.

Copyright:

© All Rights Reserved
Download as PDF, TXT or read online from Scribd
Download as pdf or txt
You are on page 1/ 3

CSC583 Artificial Intelligence Algorithms Group Assignment

CSC583 Artificial Intelligence Algorithms

GROUP ASSIGNMENT (30%)


This assignment aims to detect fraudulent bank notes (money) accurately.

For that, a set of images taken from genuine and forged banknote-like specimens is created. Features
such as wavelet variance, wavelet skewness, wavelet kurtosis, and image entropy are extracted from
the images.

DATASET
The data file BankNote.csv given is the source of information for the classification problem.

This problem has the following variables/attributes:

• variance, used as input.


• skewness, used as input.
• curtosis, used as input.
• entropy, used as input.
• class, used as the target. It can only have two values: 0 (original note) or 1 (fake).

**The dataset will be included with this assignment for your reference:

INSTRUCTIONS
Listed below are the mandatory implementation.

1. Data exploration and preparation:


In this section, you should explore the dataset to understand it through visualization and some simple
statistical analysis. You should prepare the data for model building.

Data Exploration

1) View the data types of each feature (<df>.head, <df>.tail, <df>.info) and show the descriptive
statistics for numerical feature (<df>.describe). This allows you to have a general idea of the
dataset contents. We can also check if there are any missing data, and if yes, if they are severe
or not.

2) Briefly report on the properties of the data such as number of rows, features/attributes

Data Preparation

1) For features with missing data, find and show the missing data and then replace them with
the appropriate values.

2) You may implement any other useful data cleaning or transformation operations that are
helpful to build a better performing model.

3) Briefly discuss the problem of having missing data in this dataset (< 70 words)
CSC583 Artificial Intelligence Algorithms Group Assignment

2. Experiment:

1) Split your dataset 70% as training and 30% for testing.

2) Briefly discuss whether the performance change if the splitting ratio changed.
(< 70 words)

3. Model Building:

1) Use neural network for building the model.

2) Briefly discuss on neural network and its application in classification problem.


(< 70 words)

4. Evaluation:

1) Generate a confusion matrix to understand how well your model is doing.

2) Briefly discuss on your model performance based on the confusion matrix generated.
(< 70 words)

SUBMISSIONS

1) You will only submit one single PDF only. Combine the cover page and notebook file
as one single PDF.

2) Cover page will be given.


3) Notebook file (.ipynb): Your notebook file should include the following:
(a) Team and member information
The first cell of your notebook must be a markdown cell that lists down the name of
your team, team members’s name and student ID and Group(A).
(b) Report
Similar to the lab practical, you should use markdown cells to describe the codes or
results in the following cell. Your report should be clear and concise. Marks are
allocated on points, not based on length. Do not put un-necessary or unimportant
details into the report.
(c) Code
This is where you write your code.
(d) Simulation result
Run the code cells and save the simulation result to include them in the submitted
notebook.
(e) Format
Save/print as pdf
CSC583 Artificial Intelligence Algorithms Group Assignment

4) Naming conventions: Do not put any spaces in the filenames. All your files must be named
using the following naming conventions:
<GroupLeader>_<Group>.<ext>

For example, a good name is as follows:

Akmal_A4CS2303A.pdf

**The template notebook file has included with this assignment for your reference:
GroupLeader_Group.ipynb.

RULES

1) Number of team members: 4 members


2) Submission deadline: WEEK 11, 7th JAN 2022 before 5 p.m. using Google Form
3) Plagiarism: You are not allowed to copy without attribution any part of someone else’s
program. The maximum penalty (100%) may be imposed on violators on both sides (the copier
and the copied).
4) Following instructions: Follow strictly the project requirement and submission instructions
specified above.

MARKINGS

Poor Average Excellent Total

Data Exploration 0-5 6-15 16-20 20

Data Preparation 0-5 6-15 16-20 20

Experiment 0-5 6-15 16-20 20

Model Building 0-5 6-15 16-20 20

Evaluation 0-5 6-15 16-20 20

100

Presentation 0-3 4-7 8-10 10

**The marks consist of coding and report writing

You might also like