The litsearchr package for R is designed to partially automate search term selection and writing search strategies for systematic reviews. It uses a keyword extraction algorithm to identify potential keywords from a sample of titles and abstracts and combines them with author- and database-tagged keywords to create a pool of possible keywords relevant to a field of study. Useful keywords in a field are identified from their importance in a keyword co-occurrence network. After keywords are grouped into concept groups, litsearchr writes Boolean searches in up to 53 languages, with stemming support for English. The searches are tested and work fully in at least 14 commonly used search databases with partial support in six additional databases.
Our intent in creating litsearchr is to make the process of designing a search strategy for systematic reviews easier for researchers in fields that lack standardized terms. By partially automating keyword selection, litsearchr reduces investigator bias in keyword selection and increases the repeatability of systematic reviews. It also reduces errors in creating database-specific searches by generating searches that work across a wide range of databases. Our hope is that litsearchr can be used to facilitate systematic reviews and contribute to automating evidence synthesis.
Check out the litsearchr website at https://elizagrames.github.io/litsearchr or install the package in R with remotes::install_github("elizagrames/litsearchr", ref="main").
litsearchr is a work in progress - any and all comments, suggestions, bugs, or questions are welcome! Please email eliza.grames@uconn.edu or open an issue at https://github.com/elizagrames/litsearchr.