INTELLIGENT SOFTWARE SYSTEMS FOR MULTI-DOMAIN DECISION SUPPORT APPLICATIONS
Tulshihar Patil, Shashank Joshi, Rahul Kadu, Fazil Hasan, Ganesh Korwar, Vinit Khetani
Abstract: Smart software systems are being more demanded to aid in the complicated decision making in heterogeneous areas like healthcare, smart infrastructure, cybersecurity, and industrial processes. This paper discusses these issues by providing a common platform of the intelligent and multi-domain decision support, the fusion of data-driven learning and algorithmic reasoning. This study aims at developing and testing a smart software architecture that can provide high-quality, justifiable, and scalable decisions in the various areas of application under dynamic environments. The suggested system combines the hybrid machine learning models, rule-based reasoning, and policy optimization based on reinforcement learning in the framework of a modular architecture. The decision fusion algorithm combines probabilistic outputs, time trends as well as contextual constraints to produce optimal recommendations. Experimental performance on representative multi-domain datasets has shown uniform improvements on decision support systems that are conventional. The incident system proposed will be able to reach the average decision accuracy of 97.6, 97.2, and 98.1 with recall and precision of 98.1 and reduced decision latency by 42 % over baseline models. The adaptive decision algorithm increases the accuracy of the policy choice by 31 percent and the false rate of the decision is 2.1. Experiments of cross-domain generalization establish strong transferability at a performance loss of below 3 %. The paper will focus on the field of architectural design, algorithmic development and quantitative validation of intelligent software systems to support multiple domains. The findings accentuate the feasibility aspect of implementing integrated intelligent decision platforms in intricate and data-intensive settings that demand making decisions.
Keywords: Intelligent software systems; Multidomain decision support; Hybrid machine learning; Decision fusion algorithms; Reinforcement learning optimization; Explainable AI
DOI: 10.24874/PES08.03.008
Recieved: 20.05.2026 Revised: 07.07.2026 Accepted: 04.08.2026
UDC:
Reads: 16 
