genderizeR (version 2.0.0)

genderize: Predicting gender for character strings.

Description

For each character string in x vector genderize use output of the findGivenNames function and returns a gender prediction for the whole character string based on possible first name terms located inside those strings.

Usage

genderize(x, genderDB, blacklist = NULL, progress = TRUE)

Arguments

x

A vector of text strings.

genderDB

A data.table output of findGivenNames function for the vector x.

blacklist

Some terms could be excluded from gender checking

progress

If TRUE (default) progress bar is displayed in the console

Value

A data table with text string, a term found in genderDB, that is finally used as a given name to predict gender, a predicted gender, number of potential gender indicators ("1" if only one term from the text string is found in genderDB).

Examples

# NOT RUN {
x = c("Winston J. Durant, ASHP past president, dies at 84", 
"Gold Badge of Honour of the DGAI Prof. Dr. med. Norbert R. Roewer Wuerzburg",
"The contribution of professor Yu.S. Martynov (1921-2008) to Russian neurology", 
"JAN BASZKIEWICZ (3 JANUARY 1930 - 27 JANUARY 2011) IN MEMORIAM", 
"Maria Sklodowska-Curie")

givenNames = findGivenNames(x)
givenNames = givenNames[count>40]
genderize(x, genderDB=givenNames, blacklist=NULL)

#                                                                             text
# 1:                            Winston J. Durant, ASHP past president, dies at 84
# 2:   Gold Badge of Honour of the DGAI Prof. Dr. med. Norbert R. Roewer Wuerzburg
# 3: The contribution of professor Yu.S. Martynov (1921-2008) to Russian neurology
# 4:                JAN BASZKIEWICZ (3 JANUARY 1930 - 27 JANUARY 2011) IN MEMORIAM
# 5:                                                        Maria Sklodowska-Curie

#    givenName gender genderIndicators
# 1:   winston   male                1
# 2:       med   male                2
# 3:        NA     NA                0
# 4:       jan   male                1
# 5:     maria female                1

# }
# NOT RUN {
# }