termExtraction function

Term extraction tool from textual fields of a manuscript

It extracts terms from a text field (abstract, title, author's keywords, etc.) of a bibliographic data frame.

termExtraction(
  M,
  Field = "TI",
  ngrams = 1,
  stemming = FALSE,
  language = "english",
  remove.numbers = TRUE,
  remove.terms = NULL,
  keep.terms = NULL,
  synonyms = NULL,
  verbose = TRUE
)

Arguments

• M: is a data frame obtained by the converting function convert2df. It is a data matrix with cases corresponding to articles and variables to Field Tag in the original WoS or SCOPUS file.

• Field: is a character object. It indicates the field tag of textual data :

  "TI"		Manuscript title
  "AB"		Manuscript abstract
  "ID"		Manuscript keywords plus
  "DE"		Manuscript author's keywords

  The default is Field = "TI".

• ngrams: is an integer between 1 and 3. It indicates the type of n-gram to extract from texts. An n-gram is a contiguous sequence of n terms. The function can extract n-grams composed by 1, 2, 3 or 4 terms. Default value is ngrams=1.

• stemming: is logical. If TRUE the Porter Stemming algorithm is applied to all extracted terms. The default is stemming = FALSE.

• language: is a character. It is the language of textual contents ("english", "german","italian","french","spanish"). The default is language="english".

• remove.numbers: is logical. If TRUE all numbers are deleted from the documents before term extraction. The default is remove.numbers = TRUE.

• remove.terms: is a character vector. It contains a list of additional terms to delete from the corpus after term extraction. The default is remove.terms = NULL.

• keep.terms: is a character vector. It contains a list of compound words "formed by two or more terms" to keep in their original form in the term extraction process. The default is keep.terms = NULL.

• synonyms: is a character vector. Each element contains a list of synonyms, separated by ";", that will be merged into a single term (the first word contained in the vector element). The default is synonyms = NULL.

• verbose: is logical. If TRUE the function prints the most frequent terms extracted from documents. The default is verbose=TRUE.

Returns

the bibliometric data frame with a new column containing terms about the field tag indicated in the argument Field.

Examples

# Example 1: Term extraction from titles

data(scientometrics, package = "bibliometrixData")

# vector of compound words
keep.terms <- c("co-citation analysis","bibliographic coupling")

# term extraction
scientometrics <- termExtraction(scientometrics, Field = "TI", ngrams = 1,
remove.numbers=TRUE, remove.terms=NULL, keep.terms=keep.terms, verbose=TRUE)

# terms extracted from the first 10 titles
scientometrics$TI_TM[1:10]

#Example 2: Term extraction from abstracts

data(scientometrics)

# term extraction
scientometrics <- termExtraction(scientometrics, Field = "AB", ngrams = 2, 
 stemming=TRUE,language="english",
 remove.numbers=TRUE, remove.terms=NULL, keep.terms=NULL, verbose=TRUE)

# terms extracted from the first abstract
scientometrics$AB_TM[1]

# Example 3: Term extraction from keywords with synonyms

data(scientometrics)

# vector of synonyms 
synonyms <- c("citation; citation analysis", "h-index; index; impact factor")

# term extraction
scientometrics <- termExtraction(scientometrics, Field = "ID", ngrams = 1,
synonyms=synonyms, verbose=TRUE)

See Also

convert2df to import and convert an WoS or SCOPUS Export file in a bibliographic data frame.

biblioAnalysis function for bibliometric analysis