phenmn_class: phenmn_class

In SchoenbergA/LinguGeo: Package for Linguistic Geographie

phenmn_class

Description

Phenomenon Classification. Semi-automatic classification of strings by expressions.

Usage

phenmn_class(
  data,
  colname,
  pat_exp,
  cl_to,
  user_class = F,
  trim = F,
  develop = T
)

Arguments

data	data.frame or SpatialPointData.frame.
colname	charater - name of column containing the strings
pat_exp	character - expressions which will be used for pattern matching
cl_to	character - classnames. Must be of same lenght as 'pat_exp'
user_class	boolean - if True the user can choose to classifie lines with multiple matches by hand.
trim	boolean - if TRUE will trim rows which have none applied class (0).
develop	boolean - if TRUE will NOT delete columns which are used to identify multiple matches. This is used to check the function.

Value

returns the dataframe with an additional column containing the classes.

Author(s)

Andreas Schönberg

Examples

# load data
csv <- read.csv(system.file("extdata","hunde.csv",package = "LinguGeo"))
# take a look
head(csv)
# reclassifie
new_class <- phenmn_class(data=csv,colname = "hunde",
                          pat_exp = c("nd|nt","ng|n.g","nn|n$")
                          ,cl_to = c("nd"   ,"ng"    ,"nn"))
# using user classification for multiple matches (user_class=TRUE)

new_class <- phenmn_class(data=csv,colname = "hunde",
                          pat_exp = c("nd|nt","ng|n.g","nn|n$")
                          ,cl_to = c("nd"   ,"ng"    ,"nn"),
                          user_class=TRUE)
# trim all rows with no class (0)
 new_class <- phenmn_class(data=csv,colname = "hunde",
                         pat_exp = c("nd|nt","ng|n.g","nn|n$")
                         ,cl_to = c("nd"   ,"ng"    ,"nn"),
                         user_class=TRUE,trim=T)

SchoenbergA/LinguGeo documentation built on March 27, 2023, 2:47 a.m.