Overview
Hello, Below is the instructions for the assignment. Detailed instruction are given in attached file.
Also, you should make a Presentation for the report. Please note that all submissions will be subject to Turnitin scrutiny for similarity and AI-generated content. The permitted similarity limit is 7% Submissions exceeding this threshold may be subject to academic penalties. You Must provide Turnitin Report for Similarity and AI percentage
Project Description
Machine Learning Applications in Business
Project Title
Business Problem Solving Using Machine Learning
Turnitin similarity: 5%
AI similarity: Less than 10%
Report Length: 3,500–4,500 words
Submission Components:
- Project Report
- Dataset, if available
- Presentation Slides
Academic Integrity Requirements
- The assignment must be the student's own work.
- All sources must be cited using APA.
- A minimum of 10 academic references (journal articles, books, industry reports, and reputable business sources) is required.
- Plagiarism, contract cheating, and unauthorised copying will be penalised according to university regulations.
Project Objective
Students are required to identify a real-world business problem, apply machine learning techniques to analyse data, apply predictive or descriptive models, and provide business recommendations.
The project should demonstrate:
- Understanding of business problems
- Data collection and preparation
- Application of machine learning techniques
- Model evaluation
- Business interpretation of results
- Team collaboration
Suggested Project Topics
You may choose one of the following app such as tawakkalna or Sehhaty or absher or anything in the below list but it should be in Saudi Arabia.
Finance
- Loan default prediction
- Credit risk assessment
- Fraud detection
- Stock price movement prediction
Marketing
- Customer segmentation
- Customer churn or attrition prediction
- Product recommendation system
- Sales forecasting
Human Resources
- Employee attrition prediction
- Recruitment analytics
- Employee performance prediction
Retail and E-commerce
- Demand forecasting
- Market Basket Analysis
- Dynamic pricing analysis
GCC Examples
- Mobile payment adoption prediction in Saudi Arabia
- Customer churn prediction for telecom companies in Saudi Arabia
- Credit card fraud detection in Saudi Arabia banks
- E-commerce purchase prediction in Saudi markets
Structure of the Project Report
Section 1: Introduction
Contents
- Background of the business problem
- Industry overview
- Importance of the problem
- Project objectives
- Research questions
Example
Topic: Customer Churn Prediction
Objectives:
- Identify factors causing customer churn.
- Apply ML models to predict churn.
- Recommend retention strategies.
Section 2: Literature Review
Contents
- Review of previous studies
- Review machine learning applications
- Comparing methodologies
- Identify research gaps
Example
Studies show that:
- Logistic Regression is widely used for churn prediction.
- Random Forest improves prediction accuracy.
- XGBoost performs well in large datasets.
Section 3: Dataset Description
Contents
- Dataset source
- Variables
- Number of observations
- Data quality issues
Example
Dataset:
- Telecom Customer Churn Dataset
- Number of customers
Variables:
Variable | Description |
Gender | Male/Female |
Age | Customer age |
Tenure | Years with the company |
Monthly Charges | Monthly bill |
Churn | Yes/No |
Section 4: Application of Machine Learning Model
Contents
Describe the application of at least two ML algorithms.
Examples:
Classification Problems
- Logistic Regression
- Decision Tree
- Random Forest
- Support Vector Machine
- XGBoost
Regression Problems
- Linear Regression
- Decision Tree Regression
- Random Forest Regression
Clustering Problems
- K-Means
- Hierarchical Clustering
Section 5: Evaluation of models
Contents
Evaluate models using appropriate metrics.
Classification Metrics
- Accuracy
- Precision
- Recall
Section 6: Business Insights and Recommendations
Contents
Interpret findings from a business perspective.
Example
Key Findings:
- Customers with high monthly charges are more likely to leave.
- Long-term customers have lower churn risk.
Recommendations:
- Introducing loyalty programs.
- Offer discounts to high-risk customers.
- Improve customer support services.
Section 7: Conclusion
Contents
- Summary
- Key findings
- Limitations
- Future improvements
Assessment Rubric (40 Marks)
1. Problem Identification and Business Relevance (7 Marks)
Criteria | Marks |
Clear business problem identified | 3 |
Strong justification and relevance | 3 |
Well-defined objectives | 1 |
Excellent (7 Marks)
Problems are clearly defined, realistic, and highly relevant to business.
2. Literature Review (7 Marks)
Criteria | Marks |
Relevant literature reviewed | 3 |
Critical discussion | 3 |
Research gap identified | 1 |
Excellent (7 Marks)
Use recent academic articles and critically compare studies.
3. Model Evaluation and Results (7 Marks)
Criteria | Marks |
Proper evaluation metrics | 3 |
Interpretation of results | 3 |
Discussion of limitations | 1 |
Excellent (7 Marks)
Results are clearly explained and linked to objectives.
4. Business Insights and Recommendations (7 Marks)
Criteria | Marks |
Practical business insights | 3 |
Feasible recommendations | 3 |
Strategic value | 1 |
Excellent (7 Marks)
Recommendations are realistic and actionable.
5. Report Quality and Academic Writing (3 Marks)
Criteria | Marks |
Organization and structure | 1 |
Academic writing style | 1 |
Referencing (APA 7th Edition) | 1 |
Excellent (3 Marks)
Professional report with proper formatting and citations.
6. Presentation and Viva (9 Marks)
Criteria | Marks |
Presentation quality | 3 |
Understanding of the project | 3 |
Ability to answer questions | 2 |
Equal participation | 1 |
Excellent (4 Marks)
All members demonstrate a strong understanding and contribute equally.