Introduction
We present the evolution of the LAGOS conference throughout its history, the ebb and flow in the popularity of some research areas in graph theory, and the centrality of LAGOS authors, as measured by several metrics from network science, amongst other topics.
Data mining
Evolution of paper and author numbers
The following picture shows the evolution of the number of LAGOS papers per year.
The following picture shows the evolution of the number of LAGOS authors per year.
The following picture shows the evolution of the number of co-authors per LAGOS paper. In particular, for each period and for each number k of co-authors, the plot shows the number of LAGOS papers in that period with k authors.
The following picture shows the evolution of the percentage of new distinct authors of the LAGOS papers per year.
The following picture shows the average percentage of new authors for LAGOS and other computer science conferences.
Similarity with other conferences
Given two sets A and B, the Jaccard index J(A,B) is equal to the ratio between the cardinality of their intersection and the cardinality of their union. The Sorensen-Dice index of similarity is equal to 2J(A,B)/(1+J(A,B)). The following picture shows the Sorensen-Dice index of similarity computed by comparing the set of LAGOS authors with the sets of authors for other computer science conferences.
Sex analysis
The sex of LAGOS authors has been determined mostly by querying the web service available at genderize.io (which is based on first names only), and partly by manually searching the authors on the web. The following picture shows the evolution of the percentages of male and female authors per year (the two percentages are computed with respect to the number of authors for which the sex has been assigned). The percentage of authors with no sex assigned is also shown (with respect to the total number of authors).
Topic analysis
The following pictures shows the word cloud corresponding to the words contained in the titles of LAGOS papers.
Of all the words contained in the titles of LAGOS papers in a certain time interval, the following picture shows what fraction of them are one of the most frequent 10 words.
Graph mining
The static graph (or collaboration graph) of LAGOS is an undirected graph whose nodes are the authors who presented at least one paper at LAGOS, and whose edges (a1,a2) correspond to two authors a1 and a2 who co-authored at least one paper (not necessarily presented at LAGOS). In other words, this graph is the subgraph of the DBLP graph induced by the set of LAGOS authors (for a definition of most of the notions used in this section and in the next one and for a description of the used algorithms, we refer the interested reader to the lecture notes available at https://github.com/piluc/GraphMining). The static graph contains 931 nodes and 3476 edges. Its density is equal to 0.008 and its largest connected component contains 92% of all nodes.
The following picture shows the evolution of the number of edges with respect to the number of nodes (in a log-log scale).
The following picture shows the evolution over time of the diameter of the largest connected component.
The following picture shows the evolution over time of the average distance between two nodes in the largest connected component (that is, the degrees of separation).
Centralities
The top-10 authors are the following.
- Degree: Jayme Luiz Szwarcfiter, Celina M. H. de Figueiredo, Fábio Protti, Luérbio Faria, Simone Dantas, Bruce A. Reed, Frédéric Havet, Flavia Bonomo-Braberman, Yoshiharu Kohayakawa, Uéverton S. Souza.
- Closeness: Jayme Luiz Szwarcfiter, Fábio Protti, Bruce A. Reed, Celina M. H. de Figueiredo, Luérbio Faria, Sulamita Klein, Guillermo Durán, Flavia Bonomo-Braberman, Rudini Menezes Sampaio, Frédéric Maffray.
- Betweenness: Jayme Luiz Szwarcfiter, Bruce A. Reed, Fábio Protti, Yoshiharu Kohayakawa, Celina M. H. de Figueiredo, Nelson Maculan, Guillermo Durán, Flavia Bonomo-Braberman, Cristina G. Fernandes, Martín Matamala.
Temporal graph
The temporal graph has the same set of nodes of the static graph, but the edges (a1,a2,y) correspond to two authors a1 and a2 who co-authored in year y at least one paper (not necessarily presented at LAGOS). The temporal closeness is intuitively the area covered by the plot of the temporal harmonic closeness of an author. For example, in the following figure, the plot of the temporal harmonic closeness of the top-2 authors with respect to the temporal closeness are shown.
The top-10 authors with respect to the temporal closeness are Jayme Luiz Szwarcfiter, Celina M. H. de Figueiredo, Fábio Protti, Sulamita Klein, Luérbio Faria, Mitre Costa Dourado, Simone Dantas, Frédéric Maffray, Guillermo Durán, Bruce A. Reed.