Skip to content
View sisolieri's full-sized avatar

Block or report sisolieri

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
sisolieri/README.md

Simone – Data Scientist Portfolio

LinkedIn Email

👋 About Me

I'm a data scientist with a background in mechanical engineering and a strong passion for mathematics, physics, science, and technology. After gaining initial work experience as a mechanical engineer, I realized my true interest lies in helping companies make better business decisions through data-driven insights. I believe that data is one of the greatest assets a company has, and my goal is to help businesses unlock its full potential.

Currently, I am seeking a junior data scientist or data analyst role where I can apply my skills to real-world business challenges. My experience in both engineering and data science allows me to bridge the gap between technical expertise and business impact.

🛠️ Skills

  • Languages: Python, SQL
  • Machine Learning: Scikit-learn, XGBoost, Keras, TensorFlow
  • Data Manipulation: Pandas, NumPy
  • Data Visualization: Matplotlib, Seaborn, Power BI
  • Tools: Jupyter Notebooks, Git, Docker
  • Key Areas: Data Visualization, Dynamic Dashboards Creation, Data Cleaning and Preprocessing, Time Series Forecasting, Clustering, Feature Engineering, Classification Models, Regression Models

🔥 Featured Projects

  • Description: A real-world business challenge involving sales forecasting at the product-store level using XGBoost. Clustering techniques were used to group products and stores to optimize marketing and inventory management strategies.
  • Technologies: Python, XGBoost, KMeans, Scikit-learn, Pandas, Power BI
  • Key Results: Implemented a sales forecasting model with an RMSE of 7-10 units, identified key clusters for more efficient stock management.
  • Description: This project analyzes public grants issued by the Ayuntamiento of Barcelona, focusing on the distribution of funds across various sectors and forecasting future grant amounts. The forecasting model was designed to predict the total grants for September 2024 based on historical data.
  • Technologies: Python, Pandas, XGBoost, CatBoost, ARIMA, Matplotlib, Seaborn
  • Key Results:
    • XGBoost was chosen as the best-performing model, with an RMSE of 441,278.
    • The analysis revealed that sectors like Culture and Sports have been prioritized since 2020, while sectors such as Urban Development and Social Rights have seen decreasing investments.
    • The predictive model is being used as a tool for financial planning, providing valuable insights into the allocation of public resources.
  • Description: A deep learning project that focuses on image classification using Convolutional Neural Networks (CNNs). Various structural modifications and techniques like data augmentation were applied to improve model performance.
  • Technologies: Python, Keras, TensorFlow, CNN, Data Augmentation
  • Key Results: Achieved 90.7% accuracy after implementing data augmentation techniques.

📬 Contact Me

Pinned Loading

  1. Prova_DS_SaloOcupacio2024 Prova_DS_SaloOcupacio2024 Public

    Admission challenge to Hackató Saló Ocupació by Barcelona activa

    Jupyter Notebook

  2. ds-market-data-science-final-project ds-market-data-science-final-project Public

    Final project for my Master's in Data Science. It includes Business Intelligence with Power BI, KMeans clustering of products and stores, and multivariate sales forecasting using machine learning m…

    Jupyter Notebook

  3. CNN-CIFAR10-classification CNN-CIFAR10-classification Public

    Classification of CIFAR-10 images using a CNN model as part of my Master's in Data Science. 12 experiments were conducted to improve accuracy, achieving 90.7% with data augmentation. Future work f…

    Jupyter Notebook 1