K-ML – Portfolio

Machine learning

Welcome to my portfolio! where data meets creativity & drive business growth

Welcome to my Machine Learning hub! I am a results-driven professional with a robust skill set in machine learning, encompassing predictive modeling, data preprocessing, and algorithm development. My ability to translate complex data into actionable insights and my commitment to staying at the forefront of ML advancements make me an ideal candidate. Let’s elevate your projects with my expertise in creating intelligent, data-driven solutions. Ready to contribute effectively from day one, I am your go-to ML specialist.

Enhancing Breast Cancer Diagnosis with SVM

This project applies a data-driven approach to breast cancer classification using Support Vector Machines (SVM) in Python, with a focus on achieving high diagnostic accuracy. It emphasizes the importance of hyperparameter tuning and data normalization in enhancing the performance of machine learning models. The project utilizes Python, Scikit-learn, Matplotlib, and Seaborn within Jupyter Notebooks. The optimized SVM model demonstrates a notable accuracy of 94.74%, illustrating the benefits of meticulous model optimization in medical research analytics.

Optimizing Breast Cancer Prediction with Cross-Validation

This project explores the use of linear regression for breast cancer prediction, with a strong emphasis on the role of cross-validation in enhancing model accuracy and generalizability. It involves a detailed analysis of breast cancer FNA sample data to predict malignancy, assessing the model’s performance using error metrics such as MSE, MAE, and RMSE. The project demonstrates that cross-validation is a crucial step in the predictive modeling process, ensuring the robustness and reliability of the model in clinical settings.

Portugal2019-ElectionAnalysis-ML

This project, titled “Portugal2019-ElectionAnalysis-ML,” involves a detailed machine learning analysis of the 2019 Portugal legislative election results. Utilizing a dataset from the UCI Machine Learning Repository, it encompasses over 10,000 observations and 30 attributes. The project includes data preprocessing, exploratory analysis, feature engineering, and the development and evaluation of six different regression models. Key findings show that Decision Tree and Random Forest models performed exceptionally well. The project, open for contributions and under the MIT License, aims to predict electoral outcomes with high accuracy.

Let's Work Together

Ready to collaborate on data-driven projects? I’m eager to bring my expertise in Power BI, Tableau, SQL, Python and data analysis to your team. Let’s turn insights into action and create impactful solutions together. Contact me to explore exciting opportunities