Multinational list of popular first names and surnames?

Question

Is there a database containing the list of the most popular first names and surnames (with occurrence count, or at least sorted by popularity) for many nations/countries?

I need such data for the generating of sample customer database. Customers from given land should at best have the realistic names from that land.

Answer 1

There is a pretty massive list of given (first) names (~50,000), and it's carefully curated (not machine generated).

More details are available on another answer:

The best source of international human given (first) names comes from a German computer magazine. The text file has nearly 50k names that are classified by likely gender, and how popular in each country. It's carefully curated and has a friendly license (GNU Free Documentation License).

The file can be downloaded here : ftp://ftp.heise.de/pub/ct/listings/0717-182.zip (name_dict.txt contains the data).

(archive link: https://web.archive.org/web/20200414235453/ftp://ftp.heise.de/pub/ct/listings/0717-182.zip )

Answer 2

For the United States, the Census Bureau has lists of surnames from 1990 and 2000 censuses here. The US Census list for 2010 was, for a time, available on census.socrata.com but that site is no longer running. (You may be able to find it with this Wayback Machine link.) The Social Security Administration provides downloadable lists of first names by gender, year, and optionally by state, based on all Social Security registrations here.

Answer 3

Wikipedia has a category with lists of popular names linking to a handful of Wikipedia articles containing tables that should be helpful to you.

Answer 4

You can find a lot of valuable sources in wolframalpha about popular names. Bellow the table, there is an option "sources".

Here is the link

Answer 5

You can get such information from Wikidata (as mentioned by @sn3fru) with the Wikidata Query Service at https://query.wikidata.org and a formulation of a SPARQL query.

Here is a SPARQL query for number of citizens recorded with given name grouped wrt. country:

SELECT ?name ?nameLabel ?country ?countryLabel ?count
WITH {
  SELECT ?name ?country (count(?person) AS ?count) WHERE {
    ?person wdt:P735 ?name .  # First names
    ?person wdt:P27 ?country . 
  }
  GROUP BY ?name ?country
  ORDER BY DESC(?count)
  LIMIT 100
} AS %results
WHERE {
  INCLUDE %results
  SERVICE wikibase:label { bd:serviceParam wikibase:language     "[AUTO_LANGUAGE],en". }
}
ORDER BY DESC(?count)

It shows that "John" is the most popular in the US (and globally) and Jean is the most popular given/first name in France.

An example with a query for most popular surnames in Denmark is:

SELECT ?name ?nameLabel ?count
WITH {
  SELECT ?name (count(?person) AS ?count) WHERE {
    ?person wdt:P734 ?name .  
    ?person wdt:P27 wd:Q35 . 
  }
  GROUP BY ?name
  ORDER BY DESC(?count)
  LIMIT 100
} AS %results
WHERE {
  INCLUDE %results
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
}
ORDER BY DESC(?count)

In this case Hansen, Nielsen and Jensen show up on the top, see http://tinyurl.com/ybpgrk9v

Answer 6

There was an API, babynamemap, but as you can see it now redirects to the projects source on github and hasn't been maintained for two years GitHub page where the project's source used to be. Aside from that I suppose you could always scrape Wikipedia's pages. As far as the phone book idea goes it looks like there is an API for the White Pages, not sure if they accept international zip codes though. There is also the Bing PhoneBook API, here is an example of that.

Answer 7

For the purpose of generating a sample customer database, the Fake Name Generator can generate random personal details.

They have name sets from many different countries and can also generate other fake details (such as e-mail or phone number) based on a selected country (i.e. all phone numbers will have the correct length and country code).

On the order in bulk page you request a large set of fake names (and other user specified details) at once.

Answer 8

I can't find the full data set, but here is a bit on the most common names in Denmark.

• Pigenavne = Female first names.
• Drengenavne = Male first names.
• Efternavne = Surnames.
• First first name and last surname is included. All middle names are omitted.
• Surnames consisting of two hyphenated names are treated as one name.

Answer 9

Here is a dataset of first names and last names which I scraped from Wiktionary Names Appendix

https://github.com/solvenium/names-dataset

• Female_given_names.txt
• Male_given_names.txt
• Surnames.txt

Answer 10

You can use leaked Facebook name dataset with 1.6M first names and 3.5M last names: https://github.com/philipperemy/name-dataset

Answer 11

In the Netherlands the Meertens Institute keeps an accurate record of all the names of Dutch citizens. You can enter a name and (if more than 100 people have that name) it will show you all kinds of statistics on the name.

The first name interface can be found here: https://www.meertens.knaw.nl/nvb/english/
The family name interface can be found here: http://www.cbgfamilienamen.nl/nfb/index.php?taal=eng

For example, the number of Peters born each year in the Netherlands:

Note: the results are in Dutch, relevant translation key:
"is gelijk aan" = equals
"als eerste naam" = as first name
"als volgnaam" = as (one of the) middle name(s)
"verspreiding" = spread
"m/v" = male/female

Answer 12

You can use the random names generator of Namespedia to build your dataset http://namespedia.com/random-name-generator.php

Answer 13

US Census has all surnames occurring more than 100 times, sorted by frequency of occurrence here: https://www.census.gov/topics/population/genealogy/data/2010_surnames.html

US Social Security Administration has popular first names available here: https://www.ssa.gov/oact/babynames/limits.html

Answer 14

I suggest Damegender as Python Tool for this taks. This software shares very good open datasets (USA, UK, Spain, Uruguay, ...) and it is giving support to machine learning features to guess gender in names if it doesn't appear in the dataset with good results.

• https://pypi.org/project/damegender/
• https://github.com/davidam/damegender

For instance,

$ python3 main.py Susana --verbose

0 males for Susana from INE.es

95677 females for Susana from INE.es

15 males for Susana from Uruguay census

2689 females for Susana from Uruguay census

0 males for Susana from United Kingdom census

47 females for Susana from United Kingdom census

52 males for Susana from United States of America census

19669 females for Susana from United States of America census

Susana gender predicted with nltk is female

Susana gender predicted with sgd is female

Susana gender predicted with svc is female

Susana gender predicted with gaussianNB is female

Susana gender predicted with multinomialNB is female

Susana gender predicted with bernoulliNB is female

Susana gender predicted with forest is female

Susana gender predicted with tree is female

Susana gender predicted with mlp is female

Answer 15

A more interesting alternative might be to use the Wikipedia Data API and capture all the millions of people on the WikiData as well as their origin and date of birth. That way you will have a base of names from all over the world and already structured by the semantics of wikipedia.

Answer 16

I had a need for a similar dataset for some i18n name testing and couldn't find one either, so I put together sigpwned/names-by-country-dataset.

It's about 2,300 forenames and 2,300 surnames from 100+ countries, organized by country, with localized and romanized name versions, and counts when possible. The data is in easy-to-use CSV, TXT, and JSON formats, and completely free. The data is scraped from Wikipedia, so it should be (at least reasonably) reliable.

You could easily generate realistic names for individual markets with this data.

Disclaimer: in case it wasn't obvious from the rest of my response, I'm the author of the dataset.