With rapid improvements in technology every day, there are multiple strategies of archiving and digitalization that are recommended and important. Number of solutions in public administration and other organizations in economy is quickly rising because of organizations, where unity of similar processes is being lost. The main objective is a retrievement of the observed document from history, which were being digitalized by rules of already made and known strategy.
Main target is understanding the process of digitalization and a transformation of the observed sample of transcriptions from the Kingdom of Serbs, Croats and Slovenes and Kingdom of Yugoslavia. The solution was created with Python programming language with help with few of programming libraries for handling data and rules of European project ParlaMint.
A unique process was established for digitalization of scanned transcription documents that is created with a programming solution which transforms parliamentary session to a folder of digital files which are comparable to already made solutions based on the ParlaMint rules. Process of identifying content based on computer vision is explored as well, calculating an error rate and identifying segment blocks that together build a specific document.
For creation of digitalization strategy, based on already created rules, the final solution is comparable with a digital library of other digitalized parliamentary sessions that were created by other organizations which are members of the project. Final product can be treated equally, research can be applicable to many professions which can be affected in studies of history, anthropology and linguistics.
|
