In MarekProkop/names.cze: Get Information on Personal Names Used in the Czech Republic

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

names.cze

So far, R package names.cze can only do one thing: estimate the gender of people living in the Czech Republic from their names. This is a first preview, a more robust solution and more features will come later.

Installation

You can install the preview version from GitHub:

devtools::install_github("MarekProkop/names.cze")

Examples

Guess the gender of a single name:

library(names.cze)
get_sex("Josef")
get_sex("Nováková")

The get_sex function is vectorized:

get_sex(c("Jan", "Jana", "Petr", "Petra"))

Using with dplyr:

library(dplyr, quietly = TRUE)
df <- tibble(
  name = c("Jan", "Jitka", "Soňa", "Michal", "Radovan")
)
df %>% 
  mutate(sex = get_sex(name))

Using with ggplot2:

library(ggplot2, quietly = TRUE)
df %>% 
  ggplot(aes(get_sex(name))) +
  geom_bar() +
  labs(x = "gender")

Without additional parameters, the get_sex function returns the gender that is more common among name bearers in the Czech Republic. If you want more certainty, use the threshold parameter, which is a number from 0 to 1 indicating the minimum probability. If the actual probability does not exceed this value (or at least does not equal it), the function returns NA.

get_sex("Rut")
get_sex("Rut", threshold = 0.9)

If you have both first and last names stored in one variable, the get_sex function will not work because it only works with one-word names for now. In this case, send only the first name to the function, as this generally has a better predictive value. For example, you can use the word function from the stringr package.

get_sex("Josef Novák")
library(stringr, quietly = TRUE)
get_sex(word("Josef Novák"))

For debugging purposes, you can use the inspect_name function. It is not vectorized, so you can only pass one name to it.

inspect_name("Václav")

MarekProkop/names.cze documentation built on Dec. 17, 2021, 2:22 a.m.