ADAPTIVE TRUST-AWARE MULTI-CRITERIA DECISION METHOD FOR ARTIFICIAL INTELLIGENCE SYSTEMS: A CONCEPTUAL FRAMEWORK AND MATHEMATICAL MODEL

Yuldashev, Nodirbek Abdumannob o'g'li

Қўқон ДПИ. Илмий хабарлар · 2026-yil

Annotatsiya

The rapid proliferation of artificial intelligence systems across safety- critical domains has exposed fundamental limitations in existing decision-making methodologies. Current multi-criteria decision-making (MCDM) approaches operate under static assumptions, trust-aware architectures lack formal mathematical grounding, and adaptive mechanisms fail to unify both dimensions within a coherent framework. This paper introduces the Adaptive Trust-Aware Multi-Criteria Decision Method (ATMCDM), a novel conceptual framework and mathematical model designed to address these interrelated deficiencies. ATMCDM unifies trust calibration, dynamic weight adaptation, and multi- criteria optimization within a single mathematically rigorous architecture. The proposed methodology introduces six formal constructs: a multidimensional state vector, a context- sensitive trust function, a knowledge accumulation function, an adaptive weight mechanism, a composite decision function, and a dynamic reward-driven update rule. A corresponding algorithm is presented with computational complexity analysis. Through critical comparison with existing MCDM methods, trust-aware models, and adaptive decision algorithms, the paper demonstrates that ATMCDM overcomes the rigidity of conventional approaches while providing theoretical guarantees for convergence and consistency.

Maqola ma’lumotlari
MualliflarYuldashev, Nodirbek Abdumannob o'g'li
JurnalҚўқон ДПИ. Илмий хабарлар
Nashr sanasi2026-08-10
Jild8
Son07.
Betlar1069-1076
TilIngliz
DOI10.70728/c.series.ped.v08.i07.148

Kalit so‘zlar

multi-criteria decision making, trust-aware computing, adaptive systems, artificial intelligence, decision intelligence, explainable AI, Human-AI collaboration, mathematical decision model.

Ilmiy soha

Қўқон ДПИ. Илмий хабарлар jurnalidan boshqa maqolalar

Қўқон ДПИ. Илмий хабарлар — barcha maqolalar