Scribd
Upload
2 views2 pages

Machine Learning Project Report Guidelines

The Machine Learning Lab Project requires students to apply concepts from previous labs on a real-world dataset, focusing on data understanding, preprocessing, transformation, modeling, and analysis. Mandatory tasks include data exploration, missing value handling, feature scaling, encoding, outlier detection, feature transformation, and training a machine learning model. The project submission consists of a report and code notebook, with specific requirements and a marking scheme totaling 20 marks.

Uploaded by

theexamplenumber1
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
2 views2 pages

Machine Learning Project Report Guidelines

The Machine Learning Lab Project requires students to apply concepts from previous labs on a real-world dataset, focusing on data understanding, preprocessing, transformation, modeling, and analysis. Mandatory tasks include data exploration, missing value handling, feature scaling, encoding, outlier detection, feature transformation, and training a machine learning model. The project submission consists of a report and code notebook, with specific requirements and a marking scheme totaling 20 marks.

Uploaded by

theexamplenumber1
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Machine Learning Lab Project

Total Marks: 20
Submission Type: Report + Code (Notebook)
Individual Project

Objective
The objective of this project is to apply all concepts covered from Lab 1 to Lab 13 on a raw,
real-world dataset. Students will perform data understanding, preprocessing,
transformation, modeling, and analysis. The focus is on identifying problems in the data and
showing improvements after applying appropriate techniques.

Dataset Selection
• Choose any real-world raw dataset.
• Dataset must contain numerical columns, categorical columns, missing values, and
outliers.
• Dataset should have at least 300 rows.

Tasks to Perform (Mandatory)


1. Data Exploration
- Display first few rows
- Explain columns and data types

2. Missing Value Handling


- Detect missing values
- Apply two imputation techniques

3. Feature Scaling
- Apply StandardScaler and MinMaxScaler

4. Encoding
- Apply One-Hot and Label Encoding

5. Outlier Detection
- Detect using IQR
- Apply capping

6. Feature Transformation
- Apply Box-Cox or Yeo-Johnson

7. Machine Learning Model


- Train and evaluate one model
Report Requirements
• Dataset description
• Before and after analysis
• Graphs and tables
• Explanation of decisions
• Conclusion
Minimum length: 5 pages

Marking Scheme (20 Marks)


• Data Exploration: 3 marks
• Missing Values: 4 marks
• Scaling & Encoding: 4 marks
• Outliers & Transformation: 4 marks
• Model & Conclusion: 5 marks

Submission Instructions
• Submit report and notebook via email and OBE (Deadline: 3rd January 2026 12:00 pm).
• Individual work only
• Plagiarism is not allowed

You might also like