📋 Repository Overview Welcome to ML-Repo! This repository showcases a comprehensive collection of 5 end-to-end machine learning projects demonstrating proficiency in data science, predictive modeling, and business analytics. Each project is designed following industry best practices, providing complete data pipelines from exploration to deployment-ready models.
📂 Projects
-
👩🔬 Breast Cancer Diagnosis Prediction Type: Binary Classification | Algorithm: Logistic Regression Accurately predicts breast tumor diagnoses (Malignant/Benign) using digitized image features. The project applies robust preprocessing, feature analysis, and achieves 97.66% accuracy with a ROC AUC of 0.99. Key features identified: radius, texture, symmetry. Main Skills: Data cleaning, feature engineering, evaluation, interpretability. 🔗 View Project
-
📊 Employee Attrition Prediction and Analysis Type: Binary Classification | Algorithms: Logistic Regression, Random Forest, Gradient Boosting Uses HR analytics to predict employee turnover and uncover root causes through EDA and feature importance. Delivers actionable HR business insights (e.g., impact of income, overtime, management tenure) and model deployment recommendations for retention. Main Skills: EDA, model comparison, business insight generation. 🔗 View Project
-
🏥 Medical Insurance Cost Prediction Type: Regression | Algorithm: Random Forest Regressor Estimates individual medical insurance charges using health and demographic features. Achieves R² = 0.88 and reveals actionable factors (e.g., smoking, age, BMI). Contains robust data wrangling, feature engineering, and interpretable modeling pipeline. Main Skills: Feature importance analysis, regression modeling, error metrics. 🔗 View Project
-
📈 Store Sales Forecasting Type: Regression | Algorithm: Random Forest Regressor Forecasts order-level profit using retail sales data. Focuses on identifying loss drivers and actionable business recommendations (e.g., discount policies, category-level cost controls). Delivers R² = 0.82 and deep-dive profit diagnostics. Main Skills: Financial analytics, EDA, in-depth feature impact analysis. 🔗 View Project
-
📉 Telco Customer Churn Prediction Type: Classification | Algorithm: Random Forest Classifier Predicts telecom customer churn to support proactive retention. Highlights high-impact churn factors (e.g., tenure, contract type) and achieves robust results (ROC AUC: 0.83). Informs targeted retention strategies to boost CLV and reduce churn. Main Skills: Imbalanced data handling, pipeline engineering, domain translation. 🔗 View Project
🛠️ Tech Stack & Tools Languages: Python (Jupyter Notebook)
Libraries: scikit-learn, pandas, numpy, matplotlib, seaborn
Environments: Jupyter/Colab
Other: Exploratory Data Analysis, Model Evaluation, Feature Engineering
Hamdaan Peshimam
🙌 For Viewers ⭐ Star this repo if you find it helpful—your support matters!
🔎 Explore each project folder for end-to-end ML case studies, notebooks, and business-focused analytics.
📝 Open Issues or leave suggestions—constructive feedback and project discussions are always welcome.
🔁 Share this portfolio with others interested in data science or project-based learning.
📣 For Recruiters ✔️ Review my portfolio—each project demonstrates hands-on technical, business, and analytical skills.
🤝 Connect with me on LinkedIn for collaborations, full-time roles, or project opportunities.
📧 Reach out directly for interviews, project discussions, or to request additional project details.
This project is licensed under the MIT License.
🗂️ Request tailored solutions—open to creating custom analyses or dashboards for your business needs.