Digital innovations in intelligence decision-making systems: The role of machine learning

Modern intelligence systems are undergoing a profound transformation driven by digital innovations, with machine learning (ML) emerging as a key tool for enhancing decision-making processes. The rapid development of information technologies enables a shift from traditional, manual analysis methods to automated and predictive models, allowing faster processing of large volumes of heterogeneous data from various sources, including IoT sensors, satellite imagery, network traffic, and social media. This paper provides a systematic review of the theoretical foundations and practical applications of ML in intelligence systems, analyzing supervised, unsupervised, and deep learning, their advantages and limitations, as well as their impact on decision-making accuracy and efficiency. Special attention is given to the challenges of ML implementation, including technical aspects (data heterogeneity and high dimensionality), security risks (model attacks, data leaks), and ethical dilemmas (algorithmic bias, transparency, and user trust). The literature and empirical studies indicate a growing trend of integrating ML with multi-criteria decision making (MCDM) methods, human-in-the-loop models, and explainable AI (XAI), enhancing risk control and decision interpretability. Future development directions include increasing model robustness, adaptability to new threats, and synergy with other digital technologies such as blockchain and quantum analytics. The study concludes that machine learning is a tool for improving the efficiency, accuracy, and resilience of intelligence systems, while human oversight remains essential for sustainable application in a dynamic digital environment.