Implementation of a text analysis process using R for search strategy development in systematic reviews: Protocol

A systematic review is based on comprehensive information retrieval. One of the main tasks in information retrieval is the development of search strategies for systematic literature searches in bibliographic databases. This includes the identification of free-text terms and keywords. The methods for developing search strategies applied by the German health technology assessment agency, the Institute for Quality and Efficiency in Health Care (IQWiG) [1,2] are based on established methods for developing and validating study filters [3]. The text analysis software Wordstat is used for search strategy development, but the actual development of a strategy is still iterative, i.e. subjective, due to a lack of technical options.
To further refine the objective approach, the aim of our project is to develop a text analysis tool for search strategy development using the programming language “R”. To this end, we will take the following steps: The first will be to define the core functions required for search strategy development. In a second step, we will identify existing R packages that are suitable to form the basis of the functions required for our package. We will then identify gaps that cannot be filled by existing packages and create the code to provide the required functions. The next step is to compile all functions to Version 1.0 of the package, including evaluation of the R package and subsequent creation of a graphical user interface (“Shiny App”). Subsequently, further development is planned of additional functions for the R package in a version 2.0 as a proof of concept.