SZTE specialists developed BI-based decision support system backed by artificial intelligence

The new NOVA system, connected to ERP and other professional systems and repositories, helps the management of corporate groups and large companies to solve various complex financial problems and optimize tasks.

Managers can have a free text conversation with the decision support system. The system is capable of verbally explaining the elements of the collection of rules compiled by experts, the decision support areas, and the analysis methods. As a result, the artificial intelligence — if necessary — asks questions, requests additional parameters and clarifies the data, and then starts the desired analysis, the results of which are displayed and communicated verbally.

Talentis Consulting Zrt., in a consortium with the University of Szeged, won HUF 580,006,000 in support for the project "Development of a BI-based decision support system backed by artificial intelligence for corporate groups to optimize complex financial and legal tasks".

In the R&D project, carried out between May 1, 2020, and April 30, 2023, an artificial intelligence-supported, business intelligence (BI)-based decision support system (software system and management support method) was created, which collects information based on predefined rules and, after intelligent analysis, provides management support in Hungarian in various investment, financial, accounting and tendering processes.

The new decision support system and management support methodology is based on a set of rules, including financial and legal contexts, created by consultants with expertise in financial and production areas. The parameters and financial data required for the evaluation of the results are taken from a DWH data warehouse after processing specific questions asked by the company manager, i.e. the user, and then analyzed and displayed using BI tools.

It was the task of the University of Szeged to develop algorithms to support BI decisions. For this, the university's specialists created a unified company database using various data sources, on the basis of which they created a list of possible BI support questions. To form the answers to the questions, several algorithms based on machine learning and artificial intelligence were trained and the one with the best results was selected.

To support the rapidly changing financial environment, it is possible to update the data in the database or add additional data, so users always receive reliable and up-to-date advice.

Source: University of Szeged

Featured image only for illustration. Credit: Glenn Carstens-Peters / Unsplash