A Process Model for Intelligent Analysis and Normalization of Academic and Educational Data

Abstract

Every information system, business, application, network, or organization generates data in a different form and format every day. The data generated in various repositories is much more than can be analysed. Therefore, they are collected, identified, cleaned, and normalized in order to be used in the most adequate way. The research proposed a general method of preliminary preparation, which includes techniques and methods such as collecting, cleaning and normalizing data from various sources, their structural modelling to appropriate models, followed by hypothesis testing and analysis of the obtained results in order to draw conclusions from academic data. This is possible with the means of computational linguistics and with the help of Python data manipulation libraries. Experiments have been made in the field of Higher Education. The experiments show that it is possible to clean, organize, validate and model data extracted from the learning management systems of higher, secondary and primary schools in Bulgaria and use them for the purpose of drawing conclusions and extracting useful information from educational databases data.