Analysis of Publications of FEP and their Affinities

In this paper we analyse the scientific production of the researchers of the School of Economics and Management of the University of Porto (FEP). The titles of articles appearing in the bibliographic database Authenticus were used as the basis of this study. We have explored the representation in the form of a network visualised as graph, where each researcher is represented by a node. The nodes are connected by the level of similarity of the publications’ titles. To obtain this network we have used the software Affinity Miner. This software can also identify affinity groups that join researchers with similar publications. We have explored two methods for this purpose and determined that Louvain’s method produces the best results. As the original network has a rather high density of links, the visual analysis would be rather difficult if the network were not simplified. Simplification is done by removing the weakest links. Using a measure called modularity with component penalty, we determine an adequate value for the minimum weight of a link. The results obtained with this methodology are interesting. First, it can identify both central researchers of FEP and central researchers in each scientific group (e.g. in Economics, Management, etc.). The 9 affinity groups that were generated are also of interest. We note that the scientific group of Economics is divided into 3 rather cohesive subgroups. Two of those include mainly researchers from this scientific group, while the third one is a mix of researchers mainly from Economics and Management. All affinity groups are characterised by automatically generated keywords. Similar analysis can be done for all other scientific subgroups. In our opinion, the structures discovered are pertinent and provide interesting insights into how FEP is in fact organised.