Babatunde Olorisade
School of Computing and Mathematics
Automation of citation screening stage in systematic review using
text mining techniques
Systematic Review (SR) is a methodical approach to literature review that aims to
investigate and provide a complete, exhaustive and unbiased summary of the
literature or state of evidence regarding a specific research topic. Due to its impartial
and exhaustive nature, in most cases, it involves sieving through thousands of
irrelevant titles and abstracts of search results to select the few relevant ones. A
typical SR consumes a considerable amount of time and effort. In this research, we
intend to develop a system, smart enough to distinguish between the relevant and
irrelevant articles of interest (within an acceptable error range) after being shown
various examples of each class. The research will apply data analytics techniques
through text mining to aid the selection of articles relevant to a particular research
interest from the pool of search results. The result of this research will significantly
reduce the time researchers will have to expend on the selection of relevant
literature and the SR as a whole.
Postgraduate Conference 2016
Page 41