We saw before how much the OA debate dwells inside the academic world: that's why we chose to actually analyse academic publications talking about Open Access. The purpose is to find expert opinions from influential people in the debate and see what's the perception of the problems from the ones directly involved.
Our weapon of choice for the research on academic papers at this step was Web of Science (WoS): its database high compatibility and consistency about metadata over all articles made it easy to analyse and index when dealing with large numbers of articles. Our query was "open access publishing" searched into english language WoS Core Collection, which gave 811 articles as a result. We looked into abstracts in order to find potential off topic articles and all metadata was downloaded. With the help of software Vos Viewer we compiled a database cleaned from duplications and with all synonims unified; then we binary-counted all terms and expression which appear in 15+ articles. Vos Viewer assigns a relevance score based on a semantic and citation algorithm.
The visualisation can be read in no particular order. The balloons' size represents each topic relevance, while lines are links between them. Also, balloons are clustered according to their semantic field.
This visualisation tells us which topics are most relevant according to directly involved people, and it is possible to see these results are aligned with the ones we found on Reddit too. Mainly cited issues are: business model, format subscription and copyright; academic publication, development models and repositories; publication quality, strictly linked to the peer review problem, which appears to depend on the academic discipline; scientific research quality, meaning the originality of it and correspondence between data and outcome quality.
