(This is a long over-due blog post. )

When my wife and I were trying give our kiddo a name a few month ago, we held such belief: we should give him a Chinese name to repect our culture but this name should be pronounceable in America. As a data guy, I tried to solve it in the “data way”. Basically, in this post, I’m trying to find out all “Chinese names” hidden inside the US babyname dataset from the babynames package.

I would say most of the names listed here are not usable but you can get quite a few good hints, such as “Kai”, “Manning”, “Mei”, etc.. I hope it can be helpful for other Chinese parents in English environment.

library(babynames)
library(tidyverse)
library(sparkline)
knitr::opts_chunk$set(message = F, warning = F)

sparkline(0)

sample_hanzi <- read_csv("sample_hanzi.csv")

pinyin_1_word <- sample_hanzi$pinyin
pinyin_2_words <- combn(sample_hanzi$pinyin, 2) %>%
  apply(2, paste0, collapse = "")
pinyin_first_names <- c(pinyin_1_word, pinyin_2_words)

hanzi_1_word <- sample_hanzi$hanzi
hanzi_2_words <- combn(sample_hanzi$hanzi, 2) %>%
  apply(2, paste0, collapse = "")
hanzi_first_names <- c(hanzi_1_word, hanzi_2_words)

first_names <- tibble(
  pinyin = pinyin_first_names,
  hanzi = hanzi_first_names
)
chinese_babynames <- babynames %>%
  mutate(lowername = tolower(name)) %>%
  inner_join(first_names, by = c("lowername" = "pinyin")) %>%
  mutate(name = paste(name, hanzi, sep = " - ")) %>%
  select(-lowername, -hanzi)

chinese_boys <- chinese_babynames %>%
  filter(sex == "M") %>%
  mutate(year = factor(year, 1880:2017)) %>%
  arrange(name) %>%
  select(-sex, -prop) %>%
  spread(year, n, fill = 0) %>%
  gather(year, n, -name) %>%
  group_nest(name) %>%
  mutate(n = map(data, ~sum(.x$n)),
         data = map(data, ~spk_chr(.x$n))) %>%
  unnest() %>%
  arrange(desc(n)) %>%
  slice(1:200)

library(kableExtra)
kable(chinese_boys, escape = F, caption = "Boy names") %>%
  kable_styling(full_width = F) 

chinese_girls <- chinese_babynames %>%
  filter(sex == "F") %>%
  mutate(year = factor(year, 1880:2017)) %>%
  arrange(name) %>%
  select(-sex, -prop) %>%
  spread(year, n, fill = 0) %>%
  gather(year, n, -name) %>%
  group_nest(name) %>%
  mutate(n = map(data, ~sum(.x$n)),
         data = map(data, ~spk_chr(.x$n))) %>%
  unnest() %>%
  arrange(desc(n)) %>%
  slice(1:300)

kable(chinese_girls, escape = F, caption = "Girl Names") %>%
  kable_styling(full_width = F)