[Task 1] Implement Isolation Forest on the Mammography Dataset

[coderTanisha22]

Objective

Implement the Isolation Forest algorithm using scikit-learn on the Mammography dataset to understand the fundamentals of anomaly detection.

This is a learning exercise - focus on understanding the algorithm, experimenting with different settings, and analyzing the results.

---

Dataset

Mammography (ODDS Repository)

---

Requirements

Data Exploration

• Load and explore the dataset.
• Check for missing values.
• Perform any preprocessing you think is necessary.

Model Implementation

• Train an Isolation Forest model.
• Explain the important hyperparameters you choose.
• Predict anomalies.

Evaluation

Evaluate your model using the available labels.

Suggested metrics:

• Precision
• Recall
• F1-score
• ROC-AUC (optional)

Experiments

Perform a few experiments to better understand the behavior of Isolation Forest.

Some ideas:

• Try different values of contamination.
• Experiment with different values of n_estimators.
• Change max_samples.
• Try the model with and without feature scaling.
• Compare with another anomaly detection algorithm (optional).

Analysis

Write a short summary discussing:

• Your approach
• Important observations
• Which hyperparameters had the biggest impact
• Challenges faced (if any)

---

Deliverables

• Clean and well-documented code
• Jupyter Notebook (.ipynb)
• README describing your approach and findings

---

Note: Please follow the repository's Pull Request template while submitting your solution.