Data for dynamical co-word analysis of science evolution

This dataset proposes lists of multi-terms which are characteristic to domains of sciences (ex. 2000 terms about common in embryology research) and the co-occurrence of its items in the literature through time. Each datasets contains: - a list of terms (or N-grams), one term per line - a spare matrix description of their co-occurrence. On each line we have a 4-plet (period, term1Id, term2Id, number of co-occurrences). For a given period (eg. 2001) and two terms X and Y, the number of co-occurrences is the number of times X and Y appear altogether in an abstract or a title of a reference corpora. In this study, the two corpora are the Thomson ISI Web of Science and MedLine. The name of the corpora is given in the filename.