CODEWITHSUNDEEP
rpos-taggerterm-vectors
Hu_Ca
Hu_Ca

Reputation: 47

term list / term vector pos-tagging in R

I have a .csv file with only one column containing 1000 rows. Each row contains a word (bag-of-words model). Now I want to find out for each word whether it is a noun, verb, adjective etc. .I would like to have a second column (with 1000 rows), each containing the information (noun or verb) belongig to the word in column 1.

I already have imported the csv into R. But what do I have to do now?

[Here is an example. I have these words and I want to find out whether it is a noun verb etc] [enter image description here

Upvotes: 1

Views: 209

Answers (2)

clemens
clemens

Reputation: 6813

You could use spacyr which is an R Wrapper to the Python package spaCy.

Note: you will have to 

library(spacyr)

spacy_initialize(python_executable = '/path/to/python')

Then for your terms:

Terms <- data.frame(Term = c("unit",
                    "determine",
                    "generate",
                    "digital",
                    "mount",
                    "control",
                    "position",
                    "input",
                    "output",
                    "user"), stringsAsFactors = FALSE)

Use the function spacy_parse() to tag your terms and add them to your dataframe:

Terms$POS_TAG <- spacy_parse(Terms$Term)$pos

The result is:

        Term POS_TAG
1       unit    NOUN
2  determine    VERB
3   generate    VERB
4    digital     ADJ
5      mount    VERB
6    control    NOUN
7   position    NOUN
8      input    NOUN
9     output    NOUN
10      user    NOUN

Upvotes: 0

phiver
phiver

Reputation: 23608

There are multiple options, but you could use udpipe for this. The 

terms <- data.frame(term = c("unit", "determine", "generate", "digital", "mount", "control", "position", "input", "output", "user"),
                    stringsAsFactors = FALSE)

library(udpipe)

# check if model is already downloaded. 
if (file.exists("english-ud-2.0-170801.udpipe")) 
  ud_model <- udpipe_load_model(file = "english-ud-2.0-170801.udpipe") else {
    ud_model <- udpipe_download_model(language = "english")
    ud_model <- udpipe_load_model(ud_model$file_model)
  }


# no need for parsing as this data only contains single words.
t <- udpipe_annotate(ud_model, terms$term, parser = "none")
t <- as.data.frame(t)
terms$POSTAG <- t$upos

terms
        term POSTAG
1       unit   NOUN
2  determine   VERB
3   generate   VERB
4    digital    ADJ
5      mount   NOUN
6    control   NOUN
7   position   NOUN
8      input   NOUN
9     output   NOUN
10      user   NOUN

Upvotes: 1

Related Questions