In my thesis I analyse, examine the governmental task structure, the available database of the task cadastre. The data is collected from the Hungarian legal regulation and strategic documents.
Nowadays a very popular, developing area is text mining. It can help the automated update of the textual information of task cadastre and the available sources.
For the reform of public administration operation the development of task cadastre is essential: this database can be the fundamental part of a great management information system providing data for deregulation (elimination of the existing contradictions, content-based legal cleansing) Connecting the Organisational database or the systems of Human Resource Management (HRM) the cooperating organisations and responsible departments can be analysed. The HR strategy can rely on the analytical information of required competencies, forecasting the needs for special training, etc.
I defined use cases supporting deregulation, general queries of overview focusing on the task cadastre. As the first important cornerstone of the great management information system I defined the goal of the text mining application to provide up-to-date task cadastre database relying on the continuously changing legal regulation instead of the static, outdated snapshot of manual systemization in 2012. I planned the workflow, created the database, separated a preprocessing phase with data cleansing and parsing from the updating system running automated processed supervised by human resources. I implemented the client-server modelled updating system to provide an easy and safe (logged and supervised) interface for the responsible officers (not to manipulate the database directly).