Advanced Problem-Solving and Applications

View at publisher Summarise with VIRAT
Authors: Kuldeep Singh Kaswan, Jagjit Singh Dhatterwal, Anand Nayyar
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Personalized Virtual Assistants (PVAs) are deeply integrated into moving human-computer interactions, which can make these agents flexible and context-aware as they need to have complex decision-making ability to deliver tailored responses. In this chapter, the Soar cognitive architecture’s application to PVAs is examined through an in-depth numerical analysis that covers a breadth of topics such as knowledge representation, problem-solving efficiency, and system adaptability. The rule-based processing system in Soar, chunking, and reinforcement learning capabilities facilitate its ability to learn dynamically and make real-time decisions. Soar’s computation efficiency is evaluated through the investigation of critical performance measures such as response latency, rule execution efficiency, scalability and memory usage. This chapter further discusses how Soar has been combined into a multimodal data processing system and compares the performance of its task-level context retention with transformer-based models. The numerical evaluation demonstrates Soar’s energy efficiency, robustness for high-concurrency conditions, and ability to leverage reinforcement learning for optimal decision policies. In addition, ethical decision-making simulations and real-world case studies offer valuable lessons on using generative AI and adaptability for customer service implementations. These results provide evidence that, whereas Soar is efficient in inexact reasoning, it shows numerous weaknesses in vague inquiries, which could be easily improved with hybrid AI integrations. This work highlights Soar as a scalable, low-energy, cognitive-driven framework for next-generation PVAs.

Fields of Study & Subjects

AI-based Problem Solving and Planning Psychiatry Mental Health Neuroscience Artificial Intelligence Applications

Publication details

Journal / source
CRC Press eBooks
Pages
211–223
Publication date
26 March 2026
Language
EN
Document type
Book chapter