Scribd
Upload
61 views2 pages

Project Assignment.2024

This document outlines an assessment task for a data mining course requiring students to build a data mining framework from scratch. The framework must include one clustering/classification method and one sequence mining method. Students will complete four steps: 1) pre-processing and analyzing a dataset, 2) implementing a classification algorithm using Weka, 3) implementing another algorithm, and 4) evaluating models and writing a report. The project is worth 20% of the student's final mark and is due by May 26th.

Uploaded by

tin nguyen
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
61 views2 pages

Project Assignment.2024

This document outlines an assessment task for a data mining course requiring students to build a data mining framework from scratch. The framework must include one clustering/classification method and one sequence mining method. Students will complete four steps: 1) pre-processing and analyzing a dataset, 2) implementing a classification algorithm using Weka, 3) implementing another algorithm, and 4) evaluating models and writing a report. The project is worth 20% of the student's final mark and is due by May 26th.

Uploaded by

tin nguyen
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 2

IT160IU Data Mining

Spring 2024

Assessment Task: Programming


Due date: 6:00PM Sunday 26th May 2024
Weighting: 20% of final mark

Objectives
This assessment task addresses the following objectives from the subject outline:

1. Understanding the process of data mining


2. Develop skills in implementing machine learning algorithms.
3. Develop skills in processing and transforming datasets.
4. Develop skills in evaluating the data mining methods.

Overview
In this assignment, you will build a data mining framework from scratch. This framework
contains one clustering/classification method and one sequence mining method. The
suggested dataset is one of the following resources:
- https://www.kaggle.com/datasets/podsyp/production-quality
- https://www.kaggle.com/datasets/fedesoriano/wind-speed-prediction-dataset
- https://www.kaggle.com/datasets/jmmvutu/summer-products-and-sales-in-
ecommerce-wish
Given a certain training dataset following a specific data format, your framework should be
built for a classification/prediction process. You need to identify the target label for
classification/prediction. You will test your data mining framework with the real-world
dataset to evaluate the quality of the processes.
In order to build this data mining framework, you need to finish four steps as follows.
- Step 1: Identify attributes for data mining, make training and testing datasets.
- Step 2: Implement a classification/prediction algorithm (Able to refer to the Weka
library to find the best algorithm)
- Step 3: Improve the results in Step 2 by clustering, other algorithms, or analysing
data.
- Step 4: Test the built models, compare and evaluate their performance. Write a report.

Step 1: Pre-processing (20 pts)


Input: a raw dataset
Output: clean data
Process:
- Cleaning and preparing data (15pts)
- Data analysis should be involved. (5pts)

Step 2: Implement a classification/prediction algorithm (20 pts)


In this step, you will implement a classification/prediction algorithm, using the Weka library.
1
The input data needs to be converted to the ARFF format.
You have to use Weka in your code. Ref:
- https://waikato.github.io/weka-wiki/use_weka_in_your_java_code/
- https://www.youtube.com/playlist?list=PLea0WJq13cnBVfsPVNyRAus2NK-KhCuzJ

Step 3: Implement another algorithm (20 pts)


In this step, you will implement another algorithm for classification/prediction or another
solution improving the above results.

Step 4: Model Evaluation and Report (20 pts)


Evaluate using 10-fold cross-validation.
You should evaluate the performance of the classification/prediction models and give some
remarks on experimental results.

Hint: You should consider the run-time of building models and making predictions for
performance evaluation as well.

Your report should include introduction, body (description of processing steps and
evaluation), conclusions, and references.

Group collaboration and report presentation (20 pts)

Project Organization and Submission

Your project structure should look like this:


yourMemberIDs_Names/
|--------- report.pdf
|--------- code/
| -------- executable and program files + datasets
Your program should be able to generate binary files with corresponding classification
method names, e.g., DECISIONTREE.

Double check before you submit:


Your program should be able to handle both absolute paths and relative paths to different
training and test files. Do not assume that the files are located in the same folder of your
programs.
Now you can zip your assignment folder and submit it on the IU Blackboard.
Congratulations, you just finished the programming assignment of IT160IU!

You might also like