github.com/fannykassapian/olympic_games_viz
2019/06/17
The Olympic Games are considered the world's foremost sports competition with more than 200 nations participating. The modern Olympic Games took place for the first time in Athens, in 1896.
This notebook explores a dataset that was made available by Randi Griffin on Kaggle. It contains data about each and every athlete that has competed at the Olympics since 1896, up until the Rio 2016 summer games.
fannykassapian.github.io/iss.html
2019/06/07
The International Space Station (ISS) is the largest human-made body in low Earth orbit. It has been inhabited continously since November 2000 and NASA is considering allowing tourists to visit it!
This map tracks the real time position of the ISS. It is updated every 15 seconds thanks to the Open Notify API. The name of the country, sea or ocean above which the ISS is located is retrieved via the GeoNames API.
Data: The Open Notify API is a collection of NASA and space APIs for public use, developped by Nathan Bergey.
The GeoNames geographical database covers all countries and contains over eleven million placenames that can be downloaded or accessed via their web services.
tailoredpath.com/transition.html
2019/05/28
Lifelong learning is on the rise and career transitions are becoming increasingly common.
This interactive web app allows the user to compare their current education level, skills and knowledge to that expected in their dream job.
The app accesses the data via an API built with Python Flask. The charts are built in JavaScript with chart.js.
Data: The O*NET program is the primary source of occupational information in the US. For each job, O*NET rates the importance and level of all skills, knowledge and abilities, along with the required levels of education and experience. This data is available on their resource center.
Project page (web app)
Notebook (data wrangling and Python functions)
github.com/fannykassapian/metro-traffic-data-analysis
2019/05/20
The Paris metro is one of the densest and busiest metro systems in the world. In 2018, 971,366,669 people checked-in to its network of 302 stations.
In these notebooks, I explored, cleaned and analyzed data about the Paris metro traffic in 2018. I produced a number of "traditional" visualizations with matplotlib, and more elaborate visualizations with the iPyLeaflet library and Kepler.gl (web-based application for visual exploration of large-scale geolocation data sets developped by Uber).
Data:
Île-de-France Mobilités, formerly STIF, is the organisation authority that controls and coordinates the different transport companies operating in the Paris-area public transport network and Île-de-France region.
In its open data portal, Île-de-France Mobilités provides the daily traffic and hourly profiles of all stops (for both its rail & surface network, since 2015), along with their geographical coordinates and information about each stop.
Notebooks & data description
Analysis & visualization (final notebook)
2018-2019
The singular format of a map allows the non-linear reading of data, thus encouraging its serendipitous exploration. Why not use it for non-geographical data, as well?
This web map uses the Uniform Manifold Approximation and Projection (UMAP) algorithm to reduce dimensionality without compromising the richness of the initial information.
Each marker's position is computed using the skills & knowledge necessary to perform a job as dimensions, and then projected in a two-dimensional space using UMAP.
The map was built with Leaflet.js, an open-source JavaScript library for interactive maps.
Data: The O*NET program is the primary source of occupational information in the US. For each job, O*NET rates the importance and level of all skills, knowledge and abilities, along with the required levels of education and experience. This data is available on their resource center.
Every year, the US Bureau of Labor Statistics (along with the State Workforce Agencies) conducts the Occupational Employment Statistics (OES) survey. The survey measures employment levels and wages accross the US. All the data, since 1988, is available here. Among its occupational data, the BLS provides its 2026 employment projections for all occupations.
Project page (map)
Article on dimensionality reduction (Towards Data Science)
Notebooks on dimensionality reduction (GitHub)
Article on how to build a non-geographical map with Leaflet (Towards Data Science)
Pen on how to build an interactive map with a side panel (CodePen)