A Visual Text Mining approach for Systematic Reviews

A Systematic Mapping on the use of Visual Data Mining to Support the Conduct of Systematic Literature Reviews

“…There is also an opportunity to extend Malheiros et al's work [8] which makes use of VDM techniques to support study selection in the SLR process as applied in the SE domain. New visualization techniques could be proposed and used together with the document map previously suggested by the authors.…”

“…A comprehensive analysis of these tools is given in the references abovementioned. One of these tools -PEx [8], is specific to be used in the SE area. A summary of tools that extract information from textual documents (e.g., UMLS-based spelling error correction, REgenstrief eXtraction; CliniViewer; NLM's MetaMap and MedLEE, among others) is given in [27].…”

“…Some VDM tools that can be used to support activities part of an SLR process are: Projection Explorer -PEx (http://infoserver.lcad.icmc.usp.br/infovis2/PEx) [8]; Tétralogie platform [21]; Spotfire-DXP (http://www.tibco.com) [23]; PDQ_MED [28]; Muscorian, a biological text mining system [32]; Semantic Features in Text -SENT [35]; Assert (http://www.nactem.ac.uk/assert) [36] and SYstematic Review Information Automated Collection System -SYRIAC [37]. A comprehensive analysis of these tools is given in the references abovementioned.…”

“…Malheiros et al [8] highlighted that VDM might significantly improve the SLR process. They found that the VDM approach accelerated the selection process and also improved the precision of the selection of relevant studies.…”

Using Visual Text Mining to Support the Study Selection Activity in Systematic Literature Reviews

“…As an extension of this idea, Visual Text Mining (VTM) is the association of mining algorithms and information visualization techniques that support visualization and interactive data exploration [6]. In recent years, there has been an increasing interest in the use of visualization techniques as supporting tools for SLRs [7,8,9,10]. This interest is motivated by the fact that humans present strong visual processing abilities; therefore visualbased techniques make use of these abilities, by using the human system to support knowledge discovery [4] i.e., in the SLR context, to discover the relevant primary studies.…”