Construisons de belles choses

Projets

Je peux parler de ce que je sais faire.
Ou je peux simplement tout te montrer..

Data Mining

Réalisation de l'ensemble du processus ETL pour Essencialia.com. Je récupère quotidiennement des données non structurées sur des sites d'annonces de biens immobiliers en utilisant Selenium et Python pour la ville d'Aracaju (Brésil).

Data is preprocessed, cleaned and loaded into a Postgres database.

Skills & tools: Selenium, Python, Postgre, SQL, BeautifulSoup, Pandas, Statistics.

Essencialia.com - Printscreen

text mining

I have been using Natural Language Processing for text mining and for extract useful information from data.

For Essencialia, I use techniques such as POS-Tagging (to extract locales from text), Sentiment Analysis (to look for news articles about violent districts) and parsing techniques to clean and process text. 

Skills & tools: Spacy, Python, Regex, scikit-learn, Machine Learning, Natural Language Processing, Aruana & Atalaia (proprietary Python libraries that I have created for text processing). 

Example of tagged text for Spacy

Data Visualization et Web Development

All the Real Estate findings are presented on Essencialia.com. I use a mix of Tableau embeddings and HTML to present the  information in a fun and interactive way. 

Data is analysed and transferred to a WordPress website using PHP. During the exploratory analysis step, I use Matplotlib, Pandas, SQL, et Python to explore data. 

Skills & Tools: Tableau, Data visualization, SQL, Pandas, Matplotlib, Python, WordPress, HTML, CSS, PHP, Javascript, Translatrepress, Elementor.

More projects

Blogging

Blogging is an essential part of communicating data and sharing knowledge. I blog and develop free content and tutorials to help other data scientists.

Development

Atalaia is a personal NLP library that I use to prototype. You can fork the latest public available version by clicking here.