Pantheon People Rankings

[mitalit]

Compare the cultural production of countries and the popularity of historical people with our rankings.

http://pantheon.media.mit.edu/rankin...-4000/2010/H15

Pantheon is a project celebrating global cultural heritage .

Pantheon is a project developed by the Macro Connections group at The MIT (USA ) Media Lab that's collecting, analyzing, and visualizing data on historical cultural popularity and production.

How does it work? It count the number of people who have a page on wikipedia?

[Creoda]

How does it work? It count the number of people who have a page on wikipedia?

https://pantheon.world/data/faq

What is Pantheon?
Pantheon is project that uses biographical data to expose patterns of human collective memory. Pantheon contains data on more than 70k biographies, which Pantheon distributes through a powerful data visualization engine centered on locations, occupations, and biographies. Pantheon’s biographical data contains information on the age, occupation, place of birth, and place of death, of historical characters with a presence in more than 15 language editions of Wikipedia. Pantheon also uses real-time data from the Wikipedia API to show the dynamics of attention received by historical characters in different Wikipedia language editions.

Pantheon’s front-end expresses the data through profiles that aggregate data for locations (cities and countries), occupations (e.g. actor, painter, astronomer, tennis player), and eras. Pantheon’s front end also contains a visualization builder, which allows users to construct custom visualizations, and a yearbook function, which provides an entertaining summary of people born each year during the twentieth century.

Who uses Pantheon?
Pantheon 1.0 was visited by over half-a-million people, including students, scholars, entertainment industry executives, and history aficionados. The data from Pantheon has been used in several academic publications. For instance, we have used it in in work connecting the centrality of languages in the network of translations with fame, in work exploring the decay of human collective memory, and in work showing how the size and composition of human collective memory changes with the introduction of communication technologies

Pantheon has also been used by people in the entertainment industry to estimate the potential market size of biographical productions. Pantheon is also a popular resource with some history teachers and scholars in the digital humanities.

How was the data from Pantheon collected?
Pantheon data was collected through a slow and painful process involving several steps. First, we identify all entities of the type biography from Wikidata. This takes a few days. Then, we use the Wikipedia API to identify all biographies that have a presence in more than 15 language editions of Wikipedia. We then run a classifier over the English text of those biographies to identify the occupation, place of birth, place of death, and gender of each biography. That classifier, called Johnny 5 in honor of the famous robot from short circuit was trained using Pantheon 1.0 data (which had been more manually curated). Johnny 5 maps locations to latitudes and longitudes and clusters them. For instance, we group people born in “The Upper East Side” or “Manhattan,” into the “New York City,” location. This classification and aggregation uses present day geographical boundaries (Pantheon does not have information on the nationality of people, but on geographical location where they were born). Then, we put the data in the front-end and search for bugs and gaps in content. We use the front-end to help clean the data, rerunning parts of the data pipeline, or sometimes, manually correcting misclassified entries. For more details on the methodology see: the Pantheon 1.0 paper in Nature’s Scientific Data.

Pantheon’s methodology is scalable, but not bulletproof. You can report data issues here.

What is HPI?
HPI stands for Historical Popularity Index. It is a simple an ad-hoc metric that aggregates information on a biography’s online popularity. HPI is based on the idea that fame needs to break multiple barriers, like those of time and language. HPI aggregates information on the age and attention received by biographies in multiple language editions of Wikipedia to provide a summary statistic of their global popularity.

HPI is currently made of five components: the “age” of a biography’s character (e.g. Jesus is more than 2,000 years old), number of Wikipedia language editions in which the biography has a presence (L), the concentration of the pageviews received by a biography across languages (L*), the stability of pageviews over time (CV), and the number of non-English pageviews received by that biography. We find that combining these metrics provides a more sensible ranking than using these metrics alone. To validate HPI, we previously showed that it correlates better with accomplishments than single metrics, when we focus on activities where individual accomplishments are measurable (e.g. Chess, Olympic Swimming, Tennis, Formula One).

While being an ad-hoc metric, HPI also attempts to correct for the internet’s English bias. By using non-English page views, and giving a premium to biographies that have a presence in multiple languages, and whose pageviews are not concentrated in only a few of them, HPI tries to move away from a ranking dominated by English pageviews.

[Colonel Frank Grimes]

Japan weirdly has the most famous sporting figures in their database by far; I can barely name any!

You must know Kazushi Sakuraba. If you haven't then your fag level has just doubled.

[JosephK]

Hmmm... all of these lists simply prove that Russia is superior to Ukraine.

Lay down your arms, Ukrainians, and study math and physics in the Donbass!

[Colonel Frank Grimes]

Hmmm... all of these lists simply prove that Russia is superior to Ukraine.

Lay down your arms, Ukrainians, and study math and physics in the Donbass!

You have to take into account population size.

Charles Murray's book on human achievement is a superior source.

EDIT: Or maybe they come to the same results. It would be interesting to check.

[Creoda]

You must know Kazushi Sakuraba. If you haven't then your fag level has just doubled.

Yes, my fag level has doubled because I don't watch men wrestling semi-naked. Such is life.

[JosephK]

You have to take into account population size.

Charles Murray's book on human achievement is a superior source.

EDIT: Or maybe they come to the same results. It would be interesting to check.

Population size, Schmopulation schmize, don't you have a flight to catch?

[Colonel Frank Grimes]

Yes, my fag level has doubled because I don't watch men wrestling semi-naked. Such is life.

You see two semi-naked men wrestling around. I see two trained killers showing their technique and skill. Your attention on them being semi-naked just tripled your fag level.

[Colonel Frank Grimes]

Population size, Schmopulation schmize, don't you have a flight to catch?

At 10pm.

100 geniuses in a population of 10,000 is notable if in contrast you have 100 geniuses in a population of 100,000.

Of course, there are factors, such as economics, culture, etc. that you may not on the surface realize has an influence but does. For example, White people across the US pretty much have the same IQ average but great accomplishments tend to come from people born in particular areas more so than others (for example, northeast of the US in contrast to southern cities even when you take into account the population size difference between the two regions).

[Sandis]

Japan weirdly has the most famous sporting figures in their database by far; I can barely name any!

I would say Japan can't be No. 1. 3 years ago i calculated all-time most famous sports people (athletes, excluding managers) and Japan was in 11th place although Western countries could have been a slight advantage. In my classification UK is divided in 4 regions:

1 United States 3821.7 Muhammad Ali, Jesse Owens, Babe Ruth
2 England 524.8 Fred Perry, James Hunt, Bobby Charlton
3 Brazil 365.1 Pelé, Ayrton Senna, Ronaldo
4 France 356.0 René Lacoste, Zinedine Zidane, Michel Platini
5 Italy 351.9 Giuseppe Meazza, Gino Bartali, Silvio Piola
6 Germany 308.2 Franz Beckenbauer, Michael Schumacher, Gerd Müller
7 Argentina 256.8 Diego Maradona, Lionel Messi, Alfredo Di Stéfano
8 Spain 226.7 Rafael Nadal, Luis Suárez, Iker Casillas
9 Canada 183.6 Gordie Howe, Wayne Gretzky, Maurice Richard
10 Netherlands 168.9 Johan Cruyff, Marco van Basten, Ruud Gullit
11 Japan 152.9 Ichiro Suzuki, Masahiko Kimura, Sadaharu Oh
12 Australia 93.8 Don Bradman, Rod Laver, Jack Brabham
13 Sweden 92.5 Björn Borg, Zlatan Ibrahimović, Gunnar Nordahl
14 Austria 90.2 Niki Lauda, Josef Bican, Matthias Sindelar
15 Russia 85.9 Lev Yashin, Valeri Kharlamov, Aleksandr Karelin
16 Finland 85.4 Paavo Nurmi, Keke Rosberg, Hannes Kolehmainen
17 Portugal 81.2 Cristiano Ronaldo, Luís Figo, Rui Costa
18 Scotland 75.1 Jackie Stewart, Jim Clark, Andy Murray
19 India 68.9 Dhyan Chand, Sachin Tendulkar, Kapil Dev
20 Belgium 65.1 Eddy Merckx, Philippe Thys, Jacky Ickx
21 Wales 56.0 Gareth Bale, Paulo Radmilovic, Billy Meredith
22 Hungary 55.1 Ferenc Puskás, László Kubala, Sándor Kocsis
23 Czech Republic 50.6 Emil Zátopek, Ivan Lendl, Petr Čech
24 Poland 49.3 Robert Lewandowski, Ernst Wilimowski, Zbigniew Boniek
25 N. Ireland 45.2 George Best, Harry Gregg, Alex Higgins
26 Uruguay 43.1 Héctor Scarone, Luis Suárez, Enzo Francescoli
27 Switzerland 39.3 Roger Federer, Clay Regazzoni, Hugo Koblet
28 Serbia 38.7 Novak Djokovic, Aleksandar Tirnanić, Blagoje Marjanović
29 Croatia 38.3 Luka Modrić, Dražen Petrović, Davor Šuker
30 Denmark 37.1 Peter Schmeichel, Poul Nielsen, Michael Laudrup

I need to improve my calculations.