Skip to content
View danieltodaDS's full-sized avatar

Block or report danieltodaDS

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
danieltodaDS/README.md

👋 Hi, I’m Daniel Toda

Contact me:

Gmail Badge

Data Science

  • Business Case: Life Safety is a health insurance company that plans to implement a cross sell strategy. For that it needs to prioritize those with higher propensity to purchase the product.
  • Solution: In this learn to rank problem, I used Gradient Boosting to classify and return a propensity score. This score was used to sort the customer base by the higher propensity to purchase and the predictions can be accessed by Google Spreadsheet
  • Results: at 20% customer base reached out, the model hit 2.89x more interested customers and reduced CAC in 35%.
  • Tools: Python, Flask, Ensemble Methods(XGBoost, Random Forest, KNN), Render Cloud, Google Scriptss
  • Business Case: A CFO of drugstore company needs to have a forecast of sales for the next six weeks, so he will know how much to invest in renovation of the stores.
  • Solution: In this time series problem, I used ensemble methods for regression tasks to predict the six weeks sales, and developed a Telegram Bot to access these predictions
  • Results: (on going)
  • Tools: Python, Flask, Ensemble Methods(XGBoost, Random Forest Regression), Render Cloud

Data Analysis

  • Business Problem: A CEO of real-state company, that consists in purchasing properties and reeseling them, asked for data team to select the best transactions that could maximize the profits
  • Solution: developed a webapp with the recomended properties(buy and sell), disposed at tables and interactive maps with the distribuition os properties
  • Results:
  • Tools: Python, Streamlit Cloud

Data Engineering

  • Business Problem: Two partners decided to enter the US fashion market, building their own e-commerce called Star Jeans but they lack of knowledge to define the product market fit.
  • Solution: Through a ETL process, the data will be scrapped from H&M e-commerce and some metrics will be presented at a dashboard to answer the business questions
  • Results: (on going)
  • Tools: Python, PowerBI, ETL

Pinned Loading

  1. Rossmann-Sales-Prediction Rossmann-Sales-Prediction Public

    Jupyter Notebook

  2. House-Rocket-Insigths-Project House-Rocket-Insigths-Project Public

    Jupyter Notebook