Portfolio

MapReduce

I created a code that does the following:

• Select pages from a PDF file and create a separate PDF file from it

• Select pages within that new PDF file and create Txt files

• Apply MapReduce functions

• Save new data in a new PDF

There is a Jupyter Notebook with a step-by-step connection to Google Drive to download the PDF file in the Kernel, and a Python file that runs everything.

The Yelp Big Data Dive: Part 1 & Part 2 

This is a Big Data project using the Yelp dataset.

This project aims to increase Yelps revenue by developing a dashboard that generates insights to target advertising campaigns.

SQL: The Art of Wrangling Data Like a Pro. Part 1

In this article,  I explain the basic SQL clauses: SELECT, FROM and WHERE.

Portfolio in Python

This Portfolio web page was made using Python, specifically Dash library. I created the whole project in English and Spanish, uploaded both of them using Heroku CLI, and link both so they can go to the English or Spanish Portfolio depending on the user. 

Chicago Bike Rides

This project was part of the Google Data Analytics course by Google. Where I analyzed the data of bike rides in the city of Chicago. With this data, I created a Tableau dashboard and a Jupyter Notebook analyzing the data and unifying 9 months worth of Bike Rides data. 

Different Ways to Connect Google Drive to a Google Colab Notebook Pt.1

In this article, I share some useful tips that I have used in my data path overtime to connect my Google Drive files into my Google Colab notebooks.

Showing Buenos Aires properties in a Map with Plotly

Buenos Aires properties from Properati data shown in a Buenos Aires map. In this article I explain how to use the Plotly library, specifically Mapbox and how to visualize some data in a map with their Latitude and Longitude, or search for that information if necessary.

Analysis of the Real Estate Market in Buenos Aires

Exploratory data analysis (EDA) and first basic Machine Learning regression model to predict the price of a property. This was one of my firsts Data Science Projects, my first academic project for the Acamica Data Science Course.

Analysis of the Real Estate Market in Buenos Aires 2.0

Graphing maps with Plotly as an in-depth EDA and analyzing any distribution between properties. More advanced Machine Learning models(XGBoost, Random Forest, Ridge, and Lasso) for predicting properties prices, and finally clustering.

Demand Forecasting for Direct Sales Business Products using Deep Learning

A forecasting model capable of predicting the necessary number of products to have in stock improves the quality of service and business dynamics on the direct sales channel. A project that was chosen as winner best overall of 107 projects of 672 people in Data Science 4 All course.

Coffee Machine Project

In this project, I worked on programming a coffee machine simulator. The machine works with typical products: coffee, milk, sugar, and plastic cups; if it runs out of something, it shows a notification. You can get three types of coffee: espresso, cappuccino, and latte. Since nothing’s for free, it also collects the money. 

Videogames Recommendation System

In this project, I worked on a Recommendation system for video games on the Steam platform. This was my 3rd Project for my Data Scientist degree at Acámica. In this project, I had data of user reviews (without rating or score) and the number of hours played per game and per user. Also, I had data information for each of the games. I created a type of rating or score using the hours played per user to create the recommendation system.