textmodel_svm: Linear SVM classifier for texts

In quanteda.textmodels: Scaling Models and Classifiers for Textual Data

Description

Fit a fast linear SVM classifier for texts, using the LiblineaR package.

Usage

textmodel_svm(
  x,
  y,
  weight = c("uniform", "docfreq", "termfreq"),
  type = 1,
  ...
)

Arguments

x	the dfm on which the model will be fit. Does not need to contain only the training documents.
y	vector of training labels associated with each document identified in train. (These will be converted to factors if not already factors.)
weight	weights for different classes for imbalanced training sets, passed to wi in LiblineaR::LiblineaR(). "uniform" uses default; "docfreq" weights by the number of training examples, and "termfreq" by the relative sizes of the training classes in terms of their total lengths in tokens.
type	argument passed to the type argument in LiblineaR::LiblineaR(); default is 1 for L2-regularized L2-loss support vector classification (dual)
...	additional arguments passed to LiblineaR::LiblineaR()

References

R. E. Fan, K. W. Chang, C. J. Hsieh, X. R. Wang, and C. J. Lin. (2008) LIBLINEAR: A Library for Large Linear Classification. Journal of Machine Learning Research 9: 1871-1874. https://www.csie.ntu.edu.tw/~cjlin/liblinear/.

See Also

LiblineaR::LiblineaR() predict.textmodel_svm()

Examples

# use party leaders for govt and opposition classes
library("quanteda")
docvars(data_corpus_irishbudget2010, "govtopp") <-
    c(rep(NA, 4), "Gov", "Opp", NA, "Opp", NA, NA, NA, NA, NA, NA)
dfmat <- dfm(tokens(data_corpus_irishbudget2010))
tmod <- textmodel_svm(dfmat, y = dfmat$govtopp)
predict(tmod)

# multiclass problem - all party leaders
tmod2 <- textmodel_svm(dfmat,
    y = c(rep(NA, 3), "SF", "FF", "FG", NA, "LAB", NA, NA, "Green", rep(NA, 3)))
predict(tmod2)

Example output

Package version: 2.1.2
Parallel computing: 1 of 1 threads used.
See https://quanteda.io for tutorials and examples.

Attaching package: ‘quanteda’

The following object is masked from ‘package:quanteda.textmodels’:

    data_dfm_lbgexample

The following object is masked from ‘package:utils’:

    View

      Lenihan, Brian (FF)      Bruton, Richard (FG)        Burton, Joan (LAB) 
                      Gov                       Opp                       Opp 
      Morgan, Arthur (SF)         Cowen, Brian (FF)          Kenny, Enda (FG) 
                      Opp                       Gov                       Opp 
    ODonnell, Kieran (FG)      Gilmore, Eamon (LAB)    Higgins, Michael (LAB) 
                      Opp                       Opp                       Opp 
      Quinn, Ruairi (LAB)     Gormley, John (Green)       Ryan, Eamon (Green) 
                      Opp                       Opp                       Opp 
    Cuffe, Ciaran (Green) OCaolain, Caoimhghin (SF) 
                      Opp                       Opp 
Levels: Gov Opp
      Lenihan, Brian (FF)      Bruton, Richard (FG)        Burton, Joan (LAB) 
                    Green                       LAB                       LAB 
      Morgan, Arthur (SF)         Cowen, Brian (FF)          Kenny, Enda (FG) 
                       SF                        FF                        FG 
    ODonnell, Kieran (FG)      Gilmore, Eamon (LAB)    Higgins, Michael (LAB) 
                      LAB                       LAB                        FG 
      Quinn, Ruairi (LAB)     Gormley, John (Green)       Ryan, Eamon (Green) 
                       FG                     Green                     Green 
    Cuffe, Ciaran (Green) OCaolain, Caoimhghin (SF) 
                    Green                        SF 
Levels: FF FG Green LAB SF

quanteda.textmodels documentation built on April 6, 2021, 9:06 a.m.