Welcome!
My name is Paulo Cuellas and this is my Portfolio webpage. Thank you for visiting it!
I am a Data Engineer an efficient and reliable professional with over eight years’ experience as a leader in data source
projects and excellent track record transforming data in results.
You may consider checking the projects listed here by clicking on the menu Projects above or just scrolling down the page. My resume
and webpage are available on the menu above as well. Once more, I would like to thank you for visiting this portfolio.
Tweet Sentiment Analysis
This is a Machine Learning application that applies Logistic Regression to classify tweets text in positive or negative
and also in political or not political. The subject is the speech of Greta Thunberg about climate change done on September 23, 2019,
on the United Nations. This was a team project. Many thanks to my friends Manoel Burgos,
Sumati Bhala and
Banafsheh Golpour.
The main technologies used by this application are:
- Python
- Flask
- jQuery, D3, and Plotly
- JSON
- Bootstrap, HTML5, and CCS3
- Tableau
- Heroku hosting the Application
- And MongoDB Atlas as database hosting
Canada Immigration from 1980 to 2013
Canadian immigration is ever-changing, and the areas from which immigration to Canada is prevalent changes every year.
This SPA displays the different areas and levels of wealth of the Countries that Canadian immigrants are coming from.
In a bar graph form and world map, it is presented the world regions and Development status of immigration to Canada, by showing the count of immigrants
associated with each. The application provides filters to select the different groups and years.
Technologies used to build this application:
- Python: Flask, flask_pymongo, and Pandas
- JS libraries: D3, Leaflet, Chart.js, and jQuery
- HTML5, CSS3, Bootstrap, and Fontawesome
- MongoDB at MongoDB Atlas
- And Railway hosting the application
Unveil IP
This Single-Page Application provides searching in a MySQL database to get IP coordinates (latitude and longitude)
and present the location in a Leaflet map. Additionally, data like country, state or province, city, zip code, and time zone also can
be visualized.
Resources used to build this application:
- Node.js
- JSON
- Leaflet
- Ajax
- jQuery, HTML5, and CCS3
- Bootstrap
- Font Awesome
- And Render hosting the SPA
- And Digital Ocean as MySQL database hosting
Tectonic plates vs seismic activity
In this Leaflet map, we can see how the earthquakes are closely connected to tectonic plates.
The visualization combines two datasets: the first one comes from the United States Geological Survey (USGS), it is a JSON URL with
Earthquakes data from the Past 7 Days', updated every 5 minutes. The second is a tectonic plates JSON file available at
GitHub.
The page provides us different options for map styles besides the tooltips with information about the
seismic activities registered.
Technologies used to build this application:
- Leaflet
- GeoJSON
- D3
- HTML5 and CSS3
Correlation between health and Income
The project is a data visualization work to present information from the U.S. Census Bureau and the Behavioral Risk Factor
Surveillance System.
The data set is based on 2014 ACS 1-year estimates
and tells us about risks facing particular demographics, including data on rates of income, obesity, poverty, smokes, lack of healthcare and age by state.
Technologies used to build this application:
- D3
- jQuery
- HTML5
- CSS3
- Bootstrap
- Font Awesome
Belly Button Biodiversity
This Single Page Application is an interactive dashboard to explore a Belly Button Biodiversity dataset, which
catalogues the microbes that colonize human navels.
The dataset reveals that a small handful of microbial species (also called operational taxonomic units, or OTUs, in the
study) were present in more than 70% of people, while the rest were relatively rare.
Technologies used to build this application:
- D3
- Plotly
- HTML5
- CSS3
- Bootstrap
- Font Awesome
Mission to Mars
This is a scrape web application that visits web sites with Mars subject, collects information and pictures, and displays them in a single HTML page.
Links to the web sites scraped are available on this SPA.
Technologies used to build this application:
- Python
- Flask
- Splinter
- flask_pymongo
- JSON
- HTML
- Bootstrap
- Font Awesome
- MongoDB at MongoDB Atlas
- And Heroku as application hosting
Employees Research
The purpose of this project was evaluating an employee's CSV dataset with 300,024 records to look for any sort of data inconsistency. As a result, an incoherency between the salaries amount and job positions was detected.
Technologies used in this project:
- PostgreSQL 11
- SQLAlchemy
- Python
- Matplotlib
- And Pandas
Latitude Analysis Dashboard
This is a dashboard website to present the analysis result of how weather changes as getting closer to the equator.
We can see how the behaviour of humidity, cloudiness, temperature and wind speed is related to the latitude variation. A table with
all data used for plotting the graphs is also available to check out.
Technologies used in this project:
- HTML5
- CSS3
- jQuery
- Bootstrap
- Python Pandas
- And Font Awesome
D3 Dynamic Filter
This application deploys D3.js and DOM Manipulation to present a dataset of UFOs eye-witness reports collected
in January 2010 in the United States and Canada. A JavaScript code allows the user to set one or more filters at the same time
(city, date, state, country, and or shape) to search for UFO sightings.
Technologies used in this project:
- D3
- JSON
- Javascript
- HTML5
- CSS3
- Bootstrap
- And Font Awesome
Weather Python Analysis
The goal of this project was to build a Python Jupyter notebook to analyse the weather of 500+ cities across the
world of varying distance from the equator, and then answer the question "What is the weather like as we approach the equator?".
A series of scatter plots were created to display how the latitude variance get influence on temperature, humidity, cloudiness, and wind speed.
Technologies used in this project:
- Python
- Matplotlib
- Pandas
- requests
- citipy
- And Jupyter Notebook
Pymaceuticals Inc
The goal of this project was to analyze a dataset that contains information about a study of potential treatments
to squamous cell carcinoma (SCC), a commonly occurring form of skin cancer. During 45 days, 250 mice were treated through a variety of
drug regimes and their physiological responses were monitored throughout that time, and all data was collected.
As results of this analysis we have a Jupyter notebook with: scatter plots for how the tumor volume changes over time for each treatment,
how the number of metastatic (cancer spreading) sites changes over time for each treatment, the number of mice still alive through the
course of treatment (Survival Rate), a bar graph that compares the total % tumor volume change for each drug across the full 45 days, and
a description of three observable trends based on the data.
Technologies used in this project:
- Python
- Matplotlib
- Pandas
- Numpy
- And Jupyter Notebook