Automated content analysis: addressing the big literature challenge in ecology and evolution. Issue 11 (16th July 2016)

• Record Type: Journal Article
• Authors: Nunez‐Mir, Gabriela C.; Iannone, Basil V.; Pijanowski, Bryan C.; Kong, Ningning; Fei, Songlin
• Editors: Fitzjohn, Richard
• Is Part Of: Methods in ecology and evolution. Volume 7:Issue 11(2016:Nov.)
• Page Start: 1262
• Page End: 1272
• Publication Date: 2016-07-16
• Subjects: concept map -- literature review -- machine learning -- quantitative review -- research synthesis -- text mining -- topic modelling; Ecology -- Periodicals; Evolution -- Periodicals; 577
• DOI: 10.1111/2041-210X.12602
• Languages: English
• ISSNs: 2041-210X

Abstract

Summary: The exponential growth of scientific literature – which we call the 'big literature' phenomenon – has created great challenges in literature comprehension and synthesis. The traditional manual literature synthesis processes are often unable to take advantage of big literature due to human limitations in time and cognition, creating the need for new literature synthesis methods to address this challenge. In this paper, we discuss a highly useful literature synthesis approach, automated content analysis (ACA), which has not yet been widely adopted in the fields of ecology and evolutionary biology. ACA is a suite of machine learning tools for the qualitative and quantitative synthesis of big literature commonly used in the social sciences and in medical research. Our goal is to introduce ecologists and evolutionary biologists to ACA and illustrate its capacity to synthesize overwhelming volumes of literature. First, we provide a brief history of the ACA method and summarize the fundamental process of ACA. Next, we present two ACA studies to illustrate the utility and versatility of ACA in synthesizing ecological and evolutionary literature. Finally, we discuss how to maximize the utility and contributions of ACA, as well as potential research directions that may help to advance the use of ACA in future ecological and evolutionary research. Unlike manual methods of literature synthesis, ACA is able to process high volumes of literature at substantially shorter time spans, while helping to mitigate human biases. The overall efficiency and versatility of this method allow for a broad range of applications for literature review and synthesis, including both exploratory reviews and systematic reviews aiming to address more targeted research questions. By allowing for more extensive and comprehensive reviews of big literature, ACA has the potential to fill an important methodological gap and therefore contribute to the advancement of ecological and evolutionary research.