Page 54 - ISC PROCEEDINGS 21.4
P. 54

BIG DATA ANALYTICS AND ARTIFICIAL INTELLIGENCE IN MONITORING
                                         PUBLIC INVESTMENT EXPENDITURE
                                TOWARDS SMART PUBLIC FINANCIAL GOVERNANCE


                                 Le Thi Ngoc* , Nguyen Thi Ngoc Phuong , Nguyen Van Hai   3
                                              1
                                                                        2
                                 1  State Treasury of Vietnam, Ministry of Finance, Hanoi, Vietnam.
                                        2 National Economics University, Hanoi, Vietnam.
                                            3  Hanoi Open University, Hanoi, Vietnam.
                                                 (*E-mail: ngoclt@vst.gov.vn)


                                                         ABSTRACT
                        Public investment expenditure (PIE) from the state budget in Vietnam during the
                  2021-2025 period has been maintained at approximately 26-32% of total state budget
                  expenditure, equivalent to over 6-7% of GDP annually, reaffirming the pivotal role of
                  public investment in economic growth and infrastructure development. However,
                  practical implementation has revealed persistent challenges, including slow disbursement,
                  adjustments to total investment capital, fragmented allocation, and risks of resource
                  inefficiency and leakage.
                        In the context of national digital transformation and the need to strengthen fiscal
                  discipline, this study analyzes the potential application of Big Data analytics (BDA) and
                  Artificial Intelligence (AI) in monitoring PIE, aiming toward a Smart Public Financial
                  Management (Smart PFM) model. The research employs institutional analysis, secondary
                  data synthesis, and international comparative experience to develop an integrated
                  framework linking state budget (ST) data, public investment management, and e-
                  procurement systems.
                        The results indicate three groups of AI tools with high potential: (i) anomaly
                  detection in expenditure transactions; (ii) machine learning-based project risk scoring;
                  and (iii) disbursement forecasting to support proactive fiscal management. Effective
                  implementation requires robust data governance, algorithm transparency, and clear
                  accountability mechanisms. The study proposes the development of a Smart Treasury
                  Platform and a comprehensive public financial data governance framework aligned with
                  the requirements of the digital economy.
                        Keywords: Big data analytics; AI; public investment expenditure; public financial
                  management; fiscal oversight; digital government.


                        1. Introduction
                        Against this background deep and widespread digital transformation, with AI
                  reshaping the way governments operate, the need to enhance efficiency, transparency,
                  and integrity in the management of PIE has become more urgent than ever. From a legal
                  perspective, the management of PIE in Vietnam is grounded in the Law on the ST and the
                  Law on Public Investment, both of which emphasize principles of publicity and
                  transparency, medium-term resource allocation, and strengthened accountability. At the
                  same time, the national digital transformation orientation under Decision No. 749/QĐ-
                  TTg has set out the goal of building a digital government, digital economy, and digital
                  society, thereby establishing an important institutional framework for applying digital
                  technologies to public financial governance.


                  53
   49   50   51   52   53   54   55   56   57   58   59