Roque Leal

I am a Geographer interested in space-time modeling and its visualizations

To do this, I combine deep data learning, geographic information systems and creative programming to generate innovative and compelling products. Currently my projects focus on business intelligence using optimization tools.

I enjoy sharing innovative ideas and techniques in advanced data analysis. I invite you to contact me,

I am looking for new professional challenges with immediate availability to take off your projects 🚀

  • This map uses data from Google My Business to display information on popular hours, data on visits and the usual length of visits using anonymized data in real-time to estimate the time. →
  • 150 Million people in 330 Thousand Groups an Awesome Map using the Meetup's data to discover the social diversity of humanity: social groups, greatest influencers, popular topics, members and who are its main organizers. →
  • A quick glance at the Uber Movement Speeds map of San Francisco from millions of Uber trips using →
  • A map that combines the data of the Hotels on TripAdvisor using Location Intelligence to visualize the offer of services in Santiago de Chile. →
  • Anonymized data of Google Popular Times to analyze the impact of confinement measures on economic activity. →
  • One, two or three stars… The Michelin guide unveiled the distinctions awarded to 2,615 restaurants, find all the awarded establishments on this interactive map. →
  • Five of my projects in the Semifinal round! I invite you to discover the best stories told through the Data. →
  • Sports are spaces where many emotions converge, especially on Instagram for the #TourdeFrance2019 I have allowed myself to extract the posts and structures in a dynamic and temporary visualization of the interactions. →
  • Invited to the Mapbox Blog, a series of my favorite DataViz to share ideas and how to use data to reveal awesome Geoinsights and communicate innovative ideas. →
  • I have focused on data and opportunities to better understand smart cities, in this example, I am having fun using Waze data with Python and R scripts in Power BI that interact in Visual Mapbox to analyze traffic congestion changes of vehicles in San Francisco over the course of a week. →
  • Host to share and viralize knowledge of the main oil and gas fields in the world. →
  • In France, an exploration of users of Airbnb, it is possible to use the data of this social network to discover interesting information, a new vision to understand the real estate market and the tourism sector. →
  • Model of the Paris Metro system developed in Processing thanks to the TransitLand API and Python Pandas libraries🚇. →
  • Uber data for Mexico City in order to know the dynamics of users and drivers. →
  • This time my curiosity focuses on a global context, analyzing maritime security fears in the Strait of Hormuz, a vital route for the shipment of oil. Using Python and Marine Traffic data are the perfect combination to represent a small part of the routing of oil tankers, identifying the sensitive areas of oil transportation, actors and the timeline of recent tensions in the Gulf. →
  • Migrations through United Nations data to visualize the Origin-Destination flows using the to explore the data interactively. →
  • Find an optimal location for a new Italian restaurant in San Francisco, based on #MachineLearning algorithms. The data available at Foursquare grouped using K-means and Mapbox from within a #JupyterNotebook. →
  • On this occasion I allow myself to republish the first review of OilMap on this occasion MapBox and his team share their first impressions of my project; without your help it would not be possible to create amazing maps and visualizations. →
  • Displays a large amount of photos on a highly interactive world map. Explore photos of everyone's places with Flickr. →
  • A contemporary look at the real estate market for Latam, by analyzing the Properati website, it is possible to know the spatial dynamics of the 54 thousand properties available for Equator. Undoubtedly, a first approach to understand the real estate landscape through an interactive visualization of the data. →
  • To locate a trend in #Instagram and visualize the relationships on a map, it is possible to program an algorithm and train it with Machine Learning to identify the main posts associated with #GiroDItalia by locating users from the same place where the event occurs and discovering a new way of see this event. →
  • “People who love to eat are always the best people!” For the love of food, I decided to explore TripAdvisor. This visualization includes information about restaurants in Mexico using Python and the power of MapBox and PowerBi. →
  • The data of 54 stores belonging to an important supermarket network in Equator are analyzed, in order to know the behavior of its customers, propose a grouping model and present a scenario of transaction forecasting. →
  • Regarding Uber and its OpenSource data visualization tools, I share a visualization of the Population Employed in Chile, having as a source a Geodatamining exercise and the exploration of the Here APIs of massive geocoding. →
  • In this Jupyter Notebook, I would like to propose a long-term forecast model of the West Texas Intermediate (WTI) crude oil spot price using the US oil inventory level and based on data hosted on the Energy Information Administration website (EIA). →
  • The data available and for public use on the web are amazing, in this case to represent 20% of the companies registered in the Chilean Internal Revenue Service, using Web scraping techniques with Python and using the Here API for geocoding Massive and advanced results, a total of 120,000 locations. →
  • Telling stories with the data, in Memory of the Disappeared in Argentina, a visualization of a process that we all must remember so that it does not happen again, no doubt in occasions like this, a fact is more than a statistic. →
  • I continue exploring sporting events, this time La Vuelta España 2019, analyzing the associated tags, it is possible to have valuable observations of the interactions of users on Instagram represented in the geographical space. →
  • A collection of Instagram geoinsights associated with #lamarchamasgrande occurred in Santiago, Chile. →
  • It might be interesting to locate a headmap of tweets for EU. Using the tweepy library to interact with the Twitter API and #PowerBi services, it is possible to create a simple interface to use and discover user interactions, their behavior and different spatial trends. →
  • A colleague invited me to continue applying models with Uber data so that now I invite you to explore the city of Bogotá to discover its traffic dynamics. →
  • The phenomenon of migration in South America, causes and projections of this phenomenon are possible to visualize different scenarios using data from the UN, the World Bank and the Foundation for Peace. This time using Python and Pandas, Numpy, GeoPandas and Seaborn libraries based on the System Dynamics model for the study of population dynamics. →
  • Twitter and Spark Together, you get the most popular tweet themes according to a location and according to the volume of Twitter streaming traffic, allowing you to graph different variables of these trends. →
  • Abstraction of a granular grouping of data to reach a new meaning of spacetime. →
Roque Leal


Your message has been sent shortly, I will answer you