NOTE 2018-03-06: Many thanks to Joseph Flanagan for the pull request that fixed the bugs caused by the interface update in May 2016. Also, if you know (some) Python, take a look at some of my Python scripts, some of which use Selenium, to do some of the things that this R package does.
The purpose of this R package is to webscrape keyword-in-context (KWIC) results of some of the Brigham Young University (BYU) corpora, created by Mark Davies.
Copy and paste the following into the R console:
install.packages("devtools")
library("devtools")
install_github("ekbrown/byucorpora")
library("byucorpora")
Before using the search functions to queries the corpora, the function set_credentials must be called to set the email and associated password for the current R session. If users aren't sure which email they used when registering for the BYU corpora, they can visit corpus.byu.edu in order to figure it out. The function get_credentials returns the email currently set to be used for queries.
The function search_coca retrieves keyword-in-context results from the COCA. The argument search_terms takes a string, or a vector of strings, to be searched for, for example, search_terms = "dude" or search_terms = c("dude", "narly"). It should be noted that the COCA doesn't handle apostrophes in a way one might expect. For example, users can't directly search for "I don't know", instead, they must transform it to "I do n't know". If users have apostrophes in a search string, they should take a minute to manually search for it in a web browser to work out the exact syntax before specifying it in this function. The argument section specifies the section or sections of the COCA to search in. Options include "all" for all sections, "spok" for the spoken section (the default), "fict" for fiction, "mag" for magazine, "news" for newspaper, "acad" for academic. The argument max_type takes an integer and specifies the maximum number of unique types to return for each search string (results shown in the upper right portion of the COCA in a web browser). For example, searching for nouns with the search string "[n*]" could potentially return tens of thousands of unique types, but a user may only be interested in the 100 most frequent ones. The argument max_per_term specifies the maximum number of results to return for each search string. The argument max_total_result specifies the specifies the maximum number of total results to return. This number should be equal to or smaller than users' daily keyword-in-context view limit imposed by their category of researcher in their profile with the BYU corpora.
The function search_coha retrieves keyword-in-context results from the COHA. The argument search_terms takes a string, or a vector of strings, to be searched for, for example, search_terms = "erstwhile" or search_terms = c("erstwhile", "ere"). It should be noted that the COHA doesn't handle apostrophes in a way one might expect. For example, users can't directly search for "I don't know", instead, they must transform it to "I do n't know". If users have apostrophes in a search string, they should take a minute to manually search for it in a web browser to work out the exact syntax before specifying it in this function. The argument section specifies the section or sections of the COHA to search in, from among section = "all" for all sections, "fict" for the fiction section (the default), "mag" for magazine, "news" for newspaper, "nf" for NF Books. Also, a specific decade can be specified by the first year in the decade, for example, 1920. A combination of genres and/or decades can be specified in a vector, for example, section = c("1850", "1950") or section = c("mag", "news"). The arguments max_type, max_per_term, and max_total_result have the same functionality as in search_coca.
The function search_bnc retrieves keyword-in-context results from the BNC-BYU. The argument search_terms takes a string, or a vector of strings, to be searched for, for example, search_terms = "cheers" or search_terms = c("cheers", "bye"). It should be noted that the BNC-BYU doesn't handle apostrophes in a way one might expect. For example, users can't directly search for "I don't know", instead, they must transform it to "I do n't know". If users have apostrophes in a search string, they should take a minute to manually search for it in a web browser to work out the exact syntax before specifying it in this function. The argument section specifies the section or sections of the BNC to search in, from among section = "all" for all sections, "spok" for the spoken section (the default), "fict" for the fiction section, "mag" for magazine, "news" for newspaper, "non-acad" for non-academic, "acad" for academic, and "misc" for miscellaneous. A combination of genres can be specified in a vector, for example, section = c("mag", "news"). The arguments max_type, max_per_term, and max_total_result have the same functionality as in search_coca.
The function search_cde retrieves keyword-in-context results from the CDE. The argument search_terms takes a string, or a vector of strings, to be searched for, for example, search_terms = "mujer" or search_terms = c("mujer", "hombre"). The argument section specifies the section or sections of the CDE to search in, specified by a one-hundred-year period or periods, for example, section = "1900s" or section = c("1900s", "1800s"). Specific genres of the 1900s can be specified with "oral" for the oral section, "fict" for fiction, "news" for news, and "acad" for academic. Additionally, a combination of genres and/or one hundred year periods can be specified, for example, c("1200s", "1500s", "1800s") or c("news", "acad"). The arguments max_type, max_per_term, and max_total_result have the same functionality as in search_coca.
The output of the search functions is a data frame. This data frame, as with any data frame, can be viewed by typing the name of the data frame at the console, or by using View(dfname) (where dfname is replaced by the actual variable name). Further, the data frame can be saved to the hard drive of the user's computer (for example, with write.table or readr::write_csv) and then imported into a spreadsheet for further analysis.