1. SARA EL HABBARI - LAMSAD 2. MHAMED-AMINE SOUMIAA - Hassan first university-Settat 3. MOHAMED MANSOURIâ€‹ - National School of Applied Sciences- Berrechid, Morocco.
Every day, digitization is taking hold more and more in our activities. The use of scanned documents has become a common practice in companies and organizations of all types to promote dematerialized exchanges and facilitate document management (sorting, archiving, restitution). Despite its many advantages, digitization raises some issues in terms of integration and exploitation of scanned documents contents on data visualization platforms such as business intelligence tools. Currently, BI tools have connectors allowing the load of data from digital documents (PDF or image) but are unable to correctly read the content of these files, especially if they are scanned. In this paper, we will present solutions allowing better interpretation of data extracted from scanned documents on BI tools using artificial intelligence languages and optical character recognition tools.
OCR, Business Intelligence, Power BI, Artificial Intelligence, PDF, Image, Scanned documents, Python.