Application of methods of mathematical statistics for computational palaeography

OData support
Dr. Hosszú Gábor
Department of Electron Devices

For the mankind the grouping of the animals or plants was significant in surviving at the dawn of history (I mean here the poisonous plants or the dangerous animals), and as well as in the area of biology.

Based on some property we are able to create groups of the surrounding objects by simply observing them. But in case we want to create the groups based on too many attributes then we have to apply the methods of computational statistics and by doing so we are able to show the results graphically and effectively. Henceforth I mean the groups as clusters, and the method as cluster analysis.

The main objective of my thesis was to create a software that able to read a database, and according to the inputs it can group the writings, and maybe to show those results or ties that cannot be discovered by historians. During the work I managed to implement and use an existing algorithm known as Rock algorithm in computational paleography. This fact is a reason to extend the Matlab function libary with this newly implemented functionality.

At this point the program is able to apply wide range of the methods in phenetics to examine scripts. Furthermore it is also able to view phylogenetic trees showing possible evolutionary relations between scripts.


Please sign in to download the files of this thesis.