EPIMAP

The data

This data comes from the Clauss-Slaby's public repository that stores a "data base that records almost all Latin inscriptions". Original data consists of a bit more than 500.000 latin inscriptions. From there, we selected the subset that contains both GPS coordinates and datation. The resulting corpus contains 135.000 inscriptions. In addition to the position and datation of the inscriptions, the data also consists of the inscriptions themselves and the reported status of the owner of the inscription -which allows for filtering. The lower date has been systematically shifted of 40 years (life expectancy in ancient Rome) as a broad estimate of the time the person lived before the inscription appeared. While being more intuitive, this also allows the very well dated inscriptions to appear in the corpus for at least 40 years -which is still considered a very localized datation.

Last update of the dataset: 14/10/2020

This tool has been designed for research purposes, allowing scientists to easily and quickly explore the Clauss-Slaby dataset and perform analysis that would not had been doable manually. In this optic, a number of tunable parameters are provided in the interface. The results provided by the tool can be freely used in academical context -referencing this tool's address tool while doing so being a nice payback :)
In addition, a draft of quantitative measurements is provided in the results section. To this date:

• The list of the most frequent words corresponding to the given filter in the corpus.
• Population present in the corpus at a given time.
• The entropy of the data in the corpus. It is computed as follows: if two points on the map are closer than R km, they are linked together. This way if two groups of points are far from each other they will not be linked together: they form clusters. The normalized entropy is defined as , where is the number of clusters, the population of the cluster c, and the total population. A large entropy (maximum 1) signifies that there are several large distinct clusters, while a low entropy (minimum 0) signifies the presence of only one big cluster. This could for instance indicate a phenomena of names migration -colonisation, trading moves, cultural trend, etc. Be careful, the quality of this measure depends on the total population considered. When population is low, it might be not significative.