This package defines a few useful functions for keyword searching using the pdftools package developed by rOpenSci.
The package can be installed from CRAN directly:
To install the development version you use devtools:
There are currently two functions in this package of use to users. The first keyword_search takes a single pdf and searches for keywords from the pdf. The second keyword_directory does the same search over a directory of pdfs.
keyword_searchThe package comes with two pdf files from arXiv to use as test cases. Below is an example of using the keyword_search function.
library(pdfsearch)
file <- system.file('pdf', '1610.00147.pdf', package = 'pdfsearch')
result <- keyword_search(file,
keyword = c('measurement', 'error'),
path = TRUE)
head(result)## # A tibble: 6 x 5
## keyword page_num line_num line_text token_text
## <chr> <int> <int> <list> <list>
## 1 measurement 1 5 <chr [1]> <list [1]>
## 2 measurement 1 9 <chr [1]> <list [1]>
## 3 measurement 1 19 <chr [1]> <list [1]>
## 4 measurement 1 21 <chr [1]> <list [1]>
## 5 measurement 2 28 <chr [1]> <list [1]>
## 6 measurement 2 31 <chr [1]> <list [1]>## [[1]]
## [1] "Often in surveys, key items are subject to measurement errors. Given just the"
##
## [[2]]
## [1] "with high quality measurements of the error-prone survey items. We"The location of the keyword match, including page number and line number, the actual line of text, and a tokenized version of the text (raw text split by individual words) are returned by default.
In addition, by default the hyphenated words at the end of the text are combined with the continued word at the start of the next line. If this behavior is not of interest, set the remove_hyphen argument to FALSE.
It may be useful to extract not just the line of text that the keyword is found in, but also surrounding text to have additional context when looking at the keyword results. This can be added by using the argument surround_lines as follows:
file <- system.file('pdf', '1610.00147.pdf', package = 'pdfsearch')
result <- keyword_search(file,
keyword = c('measurement', 'error'),
path = TRUE, surround_lines = 1)
head(result)## # A tibble: 6 x 5
## keyword page_num line_num line_text token_text
## <chr> <int> <int> <list> <list>
## 1 measurement 1 5 <chr [3]> <list [3]>
## 2 measurement 1 9 <chr [3]> <list [3]>
## 3 measurement 1 19 <chr [3]> <list [3]>
## 4 measurement 1 21 <chr [3]> <list [3]>
## 5 measurement 2 28 <chr [3]> <list [3]>
## 6 measurement 2 31 <chr [3]> <list [3]>## [[1]]
## [1] "Abstract"
## [2] "Often in surveys, key items are subject to measurement errors. Given just the"
## [3] "data, it can be difficult to determine the distribution of this error process, and"
##
## [[2]]
## [1] "some settings, however, analysts have access to a data source on different individuals"
## [2] "with high quality measurements of the error-prone survey items. We"
## [3] "present a data fusion framework for leveraging this information to improve inferences"keyword_directoryThe keyword_directory function allows users to search for keywords in multiple PDF files in one function call. The same functionality from the keyword_search function can be invoked, specifically remove_hyphen and surround_lines. Below is an example of searching a single directory.
directory <- system.file('pdf', package = 'pdfsearch')
# do search over two files
directory_result <- keyword_directory(directory,
keyword = c('repeated measures', 'measurement error'),
surround_lines = 1)
head(directory_result, n = 2)## ID pdf_name keyword page_num line_num
## 1 1 1501.00450.pdf repeated measures 1 24
## 2 1 1501.00450.pdf repeated measures 2 57
## line_text
## 1 introduce more sophisticated experimental designs, specifi- only would we miss potentially beneficial effects, we may also, cally the repeated measures design, including the crossover get false confidence about lack of negative effects. Statistical, design and related variants, to increase KPI sensitivity with power increases with larger effect size, and smaller variances.
## 2 a limitation to any online experimentation platform, where within-subject variation. We also discuss practical considfast, iterations and testing many ideas can reap the most erations to repeated measures design, with variants to the, rewards. crossover design to study the carry over effect, including the
## token_text
## 1 introduce, more, sophisticated, experimental, designs, specifi, only, would, we, miss, potentially, beneficial, effects, we, may, also, cally, the, repeated, measures, design, including, the, crossover, get, false, confidence, about, lack, of, negative, effects, statistical, design, and, related, variants, to, increase, kpi, sensitivity, with, power, increases, with, larger, effect, size, and, smaller, variances
## 2 a, limitation, to, any, online, experimentation, platform, where, within, subject, variation, we, also, discuss, practical, considfast, iterations, and, testing, many, ideas, can, reap, the, most, erations, to, repeated, measures, design, with, variants, to, the, rewards, crossover, design, to, study, the, carry, over, effect, including, theA few other useful arguments are possible when searching for keywords within multiple PDF files in a directory. One is the recursive (default is FALSE), where if set to TRUE will search within subdirectories as well, the default function behavior will not venture into subdirectories. Finally, if the directory has many PDF files, testing the function first on a handful of PDF files may be desired. The number of PDF files can be limited with the argument max_search where a positive integer can be specified indicating the number of PDF files to search. For example, is max_search = 2, only the first two PDF files will be searched within the directory.
The package also has a simple Shiny app that can be called using the following command
The pdfsearch package may be most useful to those conducting research syntheses or meta-analyses. The package can allow users to search for keywords related to a research question; therefore, instead of searching the entire text of a document, specific portions of the text can be identified to be searched. This could increase the reproducibility and reduce the time needed to collect the data for the research synthesis or meta-analysis.
As an example, the package is currently being used to explore the evolution of statistical software and quantitative methods used in published social science research (https://ww2.amstat.org/meetings/jsm/2018/onlineprogram/AbstractDetails.cfm?abstractid=330777). This process involves getting PDF files from published research articles and using pdfsearch to search for specific software and quantitative methods keywords within the research articles. The results of the keyword matches will be explored using research synthesis methods. A pre-print of the paper and slides from the presentation will be posted to the GitHub repo as part of the package later this summer.