Project Overview
Personal Finance Risk Analysis Using Machine Learning
In this project, you will work with a structured dataset containing financial and demographic
attributes of 200 individuals. The primary objective is to analyze how personal financial habits
and income patterns influence overall financial risk, and to apply foundational machine learning
techniques to uncover patterns, group similar individuals, and build predictive models.
This project is designed to simulate a real-world data science workflow. You will begin by
understanding the dataset through systematic exploratory analysis, then move to pattern
discovery using unsupervised learning, and finally build a predictive model using supervised
learning techniques.
This project can be completed using Google Sheets or Microsoft Excel. However, students who
are comfortable with programming may use Python or any other analytical tool of their choice.
Project Requirements
Tools
1. Google Sheets (Recommended) or Microsoft Excel
2. Python or any other analytical tool (Optional)
Students are encouraged to use spreadsheet-based solutions to strengthen their conceptual
understanding of distance calculations, clustering mechanisms, and impurity measures.
Project Deliverables
You will be required to complete the following components:
1. Assessment Questions
Students must answer all questions provided across the four sections of the project.
2. Comprehensive Project Report
Students must submit a detailed project report that includes:
Introduction and problem statement
Dataset description
Exploratory Data Analysis with supporting tables and charts
K-Means clustering methodology and interpretation
Decision Tree model explanation and reasoning
Mathematical computations where required
Key insights derived from the analysis
Challenges encountered during implementation
Final conclusion
The report should demonstrate clarity of thought, methodological correctness, and logical
reasoning.
Important Note
For assistance in preparing the report, refer to the Guidelines Document. This document
provides a structured framework and formatting recommendations.
Students are encouraged to be creative in their analysis and visualizations, while ensuring that all
required sections are comprehensively covered.
Dataset
Link to the dataset: Click here
Note: Please make a copy of the dataset file, download it, and then perform your analysis.
Dataset Description
The dataset contains information about 200 individuals with financial and demographic
attributes. The purpose of the dataset is to analyze the relationship between personal financial
habits and financial risk classification.
Dataset Columns
1.
1. PersonID – Unique identifier assigned to each individual. This column is not used in
model training and must be excluded during algorithm implementation.
2. Age – Age of the individual (in years). This feature may influence financial behavior and
clustering patterns.
3. Gender – Gender of the individual (Male / Female). This categorical variable must be
encoded before applying machine learning algorithms.
4. MonthlyIncome – Monthly income of the individual (in currency units). Income is a
primary indicator of financial capacity and plays a significant role in classification.
5. MonthlyExpenses – Total monthly spending across all categories. This feature reflects
overall consumption behavior.
6. SavingsRate – Percentage of monthly income saved by the individual. This represents
financial discipline and long-term planning behavior.
7. DebtToIncomeRatio – Ratio of total debt to annual income. This is a critical indicator of
financial burden and repayment capacity.
8. NumCreditCards – Number of active credit cards held by the individual. This reflects
credit exposure and spending channels.
9. FinancialRisk – Target variable representing overall financial risk category. Categories
include: Low / Medium / High.
This variable will be used as the prediction target in the Decision Tree classification
model.
Structure of the Problem Statement
Accordingly, this project is divided into four structured sections:
Section 1 – Exploratory Data Analysis (EDA)
Section 2 – K-Means Clustering
Section 3 – Decision Tree Classification
Section 4 – Project Presentation
Each section builds upon the previous one, ensuring a progressive understanding of both data
and machine learning concepts.