Proximity map of Europe

[pulstar]

So I've used some socio-econo-geographic parameters (population density, net migration, infant mortality (per 1000 births), GDP per capita, literacy (%), phones (per 1000), arable land (%), crops (%), other land (%), birthrate, deathrate, agriculture, industry, service) to decide how close are different European countries. Since I wasn't sure how many similar regions there could be I've run my algorithm on multiple number of clusters.

Based on two groups:


Based on three groups:


Based on four groups:


Based on five groups:


Based on six groups:


I didn't had data for 3 Baltic countries and Montenegro.

Based on your opinion which map is the most accurate?

[Dr_Maul]

First one except for Portugal

[pulstar]

First one except for Portugal

So Malta being alongside with East Europe is ok?

EDIT (few more notions)

I can see one interesting pattern here:

1st map: poor Europeans vs rich Europeans
2nd map: poor vs average vs rich
3rd map: very poor vs poor vs rich vs very rich
4th map: very poor vs poor vs average vs rich vs very rich
5th map: very poor vs poor vs normal & a bit below average vs normal & a bit above average vs rich vs very rich

[Dr_Maul]

So Malta being alongside with East Europe is ok?

Didn't notice that, but to be honest it fits better with Georgia/Cyprus/Armenia anyway. But I think it should be blue here

Actually I think the 3 groups version is probably the most accurate
Blue: Western Europe
Yellow: Western/Civilized but less fortunate economically
Red: Balkaniggas and Slavomongols

[pulstar]

Didn't notice that, but to be honest it fits better with Georgia/Cyprus/Armenia anyway. But I think it should be blue here

Actually I think the 3 groups version is probably the most accurate
Blue: Western Europe
Yellow: Western/Civilized but less fortunate economically
Red: Balkaniggas and Slavomongols

Perhaps. What I find funny is that most of the people imagine France, Germany and UK as some sort of a "golden standard" for Western Europe and the last map shows there're like lower average in Western European terms, and definitively worse place to live than San Marino, Switzerland, Norway, Denmark or Iceland, and much worse than Luxembourg.

[Frowning Man]

The only thing I disagree with is the presence of Georgia and Armenia on these maps. They do not belong to Europe as well as the Caucasus as a whole does not belong to Europe.
Georgia (I will speak only for Georgia, since I am Georgian) is not close to any country in Europe either genetically, phenotypically, culturally, mentally or historically.
The only people who are close to the Georgians in all that I have listed are only to other Caucasians.

[brennus dux gallorum]

Interesting. I would only paint romania pink in the 5th map and anything else seems right

[gixajo]

So I've used some socio-econo-geographic parameters (population density, net migration, infant mortality (per 1000 births), GDP per capita, literacy (%), phones (per 1000), arable land (%), crops (%), other land (%), birthrate, deathrate, agriculture, industry, service) to decide how close are different European countries. Since I wasn't sure how many similar regions there could be I've run my algorithm on multiple number of clusters.



I didn't had data for 3 Baltic countries and Montenegro.

Based on your opinion which map is the most accurate?

And if you make coordinates , every coordinate a % for any parameter, and you run them in Vahaduo?

For example:

In target every country with their coordinates
C0=nameof country
C1=GDP per capita
C2=population
C3=deathrate
C4=...etc.

And in sources, classifications with the same coordinates

C0=poor country/rich country/very rich country/etc...(or colours representing each "class" of country)
C1=GDP per capita
C2=population
C3=deathrate
C4=...etc.

I didn´t check it,because I am too lazy to do it just now,but maybe it works well.

[gixajo]

Results should be 72.2%poor/20 % rich/7.8% very rich for example, and according to those results, to assign a colour to every country.

[pulstar]

Results should be 72.2%poor/20 % rich/7.8% very rich for example, and according to those results, to assign a colour to every country.

And if you make coordinates , every coordinate a % for any parameter, and you run them in Vahaduo?

For example:

In target every country with their coordinates
C0=nameof country
C1=GDP per capita
C2=population
C3=deathrate
C4=...etc.

And in sources, classifications with the same coordinates

C0=poor country/rich country/very rich country/etc...(or colours representing each "class" of country)
C1=GDP per capita
C2=population
C3=deathrate
C4=...etc.

I didn´t check it,because I am too lazy to do it just now,but maybe it works well.

Didn't try that, but I don't know if those are good labels (rich vs poor), its just a hunch feeling.

I've got the data from Kaggle.

Then I used scikit-learns standard scaler and k-means.

But yeah, I know what you want, you want to see where each country is on a scatter plot. I might make it later on if I figure out how Vahaduo works.