🤖 Machine Learning Internship – Pathline Interns (Level I)

Welcome to my Machine Learning Level I Internship portfolio!
This repository contains 3 complete projects developed during my internship with Pathline Interns, focused on solving real-world problems using foundational ML techniques.

---

📚 Internship Overview

🏢 Organization: Pathline Interns
🎓 Level: Machine Learning – Level I
👨‍💻 Role: Intern – ML Developer

Throughout this internship, I applied essential machine learning techniques such as data preprocessing, classification, evaluation, and hyperparameter tuning using Python and Scikit-learn.

---

🧪 Projects Breakdown

🔹 Task 1 – Titanic Survival Prediction (Decision Tree Classifier)

Predict survival on the Titanic based on features like age, class, fare, and sex.

• Cleaned data and handled missing values
• Applied log transformation and outlier removal
• Encoded categorical features
• Used SMOTE for class balancing
• Trained a Decision Tree Classifier with entropy
• Visualized the decision tree and analyzed accuracy

📁 Folder: Task 1 - Titanic Decision Tree Classifier/
📎 Notebook: task_1_titanic_decision_tree.ipynb

---

🔹 Task 2 – Breast Cancer Detection (SVM + Hyperparameter Tuning)

Classify tumors as malignant or benign using SVM and optimize performance.

• Encoded labels (M, B) into binary classes
• Handled feature skewness using PowerTransformer
• Scaled features with StandardScaler
• Balanced dataset using SMOTE
• Trained SVM Classifier (RBF kernel)
• Performed GridSearchCV to tune C and gamma

📁 Folder: Task 2 - Breast Cancer SVM/
📎 Notebook: task_2_breast_cancer_svm.ipynb

---

🔹 Task 3 – Book Recommender System (KNN)

Build a simple recommender system using K-Nearest Neighbors to suggest books based on user preferences and rating similarities.

• Preprocessed book metadata
• Removed duplicates and irrelevant columns
• Scaled numerical features
• Applied KNN-based content filtering
• Evaluated performance of top-N recommendations

📁 Folder: Task 3 - Book Recommender System/
📎 Notebook: task_3_books_knn.ipynb

---

⚙️ Tools & Libraries Used

• Python (Pandas, NumPy)
• Matplotlib, Seaborn
• Scikit-learn (SVM, Decision Tree, GridSearchCV, KNN)
• Imbalanced-learn (SMOTE)
• PowerTransformer, StandardScaler

---

📈 What I Learned

✅ End-to-end machine learning workflows
✅ Handling imbalanced data with SMOTE
✅ Outlier and skewness mitigation
✅ Model evaluation with precision, recall, and confusion matrices
✅ Tuning hyperparameters with GridSearchCV
✅ Visualizing and interpreting model outputs

---

💼 Let's Connect

Feel free to explore the code, clone the repo, and reach out!

---