Designing Adaptive Intelligent Assistants: A Multi-Model Learning Approach Using Logistic Regression, CRFS, Q-Learning, and K-NN

Emmanuel Imuede Oyasor

doi:10.57239/PJLSS-2025-23.2.00192

Author(s)

Emmanuel Imuede Oyasor Department of Accounting Science, Walter Sisulu University,Mthatha,South Africa

Abstract

This study introduces a robust theoretical and computational framework for designing an intelligent assistant by combining supervised and reinforcement learning techniques to enhance functionality in areas such as intent recognition, entity extraction, dialogue management, and personalized recommendations. Intent classification is handled using logistic regression, while Conditional Random Fields (CRFs) are utilized for effective sequential entity extraction. Dialogue flow is governed by Q-learning to support adaptive decision-making, and K-Nearest Neighbors (K-NN) enables responsive personalization. Experimental results indicate strong system performance, with F1-scores above 0.90 for both intent and entity tasks, and continuous improvement in dialogue decisions through learning iterations. Economically, the intelligent assistant shows promise in lowering transaction costs, boosting user satisfaction, and increasing efficiency by automating routine service interactions. The findings highlight the system's scalability, cost-effectiveness, and adaptability, with future prospects including stronger alignment with ethical AI standards and deployment in real-time customer service environments.

Publication Details

Page(s) 576-585

DOI 10.57239/PJLSS-2025-23.2.00192

Published Journal: Pakistan Journal of Life and Social Sciences, Volume: 23, Issue: 2, Year: 2025

Request the full text from PASTIC

Keywords

Intelligent Assistant Intent Classification Reinforcement Learning Personalized Recommendation Dialogue Management Economic Efficiency

References

Aguiar A. T.,Almeida M. A.,Coelho R. N. .2021 .The impact of automation on business performance: An economic analysis. Journal of Business Economics, 98(5) : 597-615.

Ahmad R.,Lu Y.,Tang J. .2025 .Learning from user cases: Enhancing personalization in dialogue systems with adaptive memory retrieval. Journal of Artificial Intelligence Research, 78(2) : 142-159.

Araujo T.,Freitas A.,Silva M. .2020 .Understanding customer experience in the digital economy. Service Science, 14(1) : 28-39.

Arora N.,Dhillon G.,Fong T. .2021 .The economics of personalized marketing: A review. Marketing Science, 40(3) : 540-560.

Arrieta A. B.,Díaz-Rodríguez N.,Ser J. D. .2020 .Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion, 58 : 82-115.

Bastani F.,Kiani A.,Bhatnagar A. .2021 .The role of natural language processing in task-oriented dialogue systems. AI & Society, 36(2) : 373-386.

Brynjolfsson E.,McAfee A. .2017 .The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. , : .

Chen Y.,Wang J.,Xu L. .2022 .Argumentative dialogue modeling for intelligent recommendation. Proceedings of the AAAI Conference on Artificial Intelligence, 36(10) : 11491-11498.

Gonzalez J.,Singh A.,Li X. .2022 .Case-based dialogue systems: A review and framework for design. Expert Systems with Applications, 197 : 116665.

González P.,Rodríguez L.,García J. .2020 .Economic impacts of artificial intelligence and automation. Journal of Economic Growth, 25(2) : 145-173.

Huang J.,Lu W.,Zhao Y. .2020 .Ontology-driven modeling for explainable recommendation systems. Knowledge-Based Systems, 192 : 105311.

Huang Z.,Benyoucef M. .2013 .From e-commerce to social commerce: A close look at design features. Electronic Commerce Research and Applications, 12(4) : 246-257.

Kim S.,Cho H. .2023 .Human-centric AI for accessibility: The case for intelligent digital assistants. ACM Transactions on Accessible Computing, 16(1) : 1-24.

Kumar S.,Singh A.,Dey L. .2019 .Optimizing intent recognition for intelligent assistants. International Journal of Human-Computer Studies, 132 : 82-99.

Lafferty J.,McCallum A.,Pereira F. .2001 .Conditional random fields: Probabilistic models for segmenting and labeling sequence data. Proceedings of ICML, : .

Liu Q.,Gao J.,Chen Z. .2021 .User modeling in task-oriented dialogue systems: A survey. ACM Computing Surveys, 54(8) : 1-37.

Mihajlov M.,Ahmed M.,Zhang Z. .2021 .Deep reinforcement learning for dialogue management: A review. AI Communications, 34(3) : 401-415.

Nguyen M.,Trinh T.,Le H. .2020 .Error correction strategies in conversational AI systems. Journal of Artificial Intelligence Research, 68(1) : 95-120.

Oliveira A.,Figueiredo D. .2019 .Cost reduction through AI in service automation. Management Science, 65(2) : 350-366.

Pereira P.,Silva M.,Lopes A. .2019 .Personalized recommendations and user engagement in AIdriven platforms. Computational Intelligence, 35(4) : 618-630.

Rojas J.,Hernandez G.,Lee S. .2020 .Deep learning models for conversational AI: Applications and insights. Journal of Artificial Intelligence & Machine Learning, 21(6) : 1093-1115.

Rojas-Barahona L. M.,Williams J.,Henderson M. .2023 .Formal approaches to dialog system verification and validation. Transactions of the Association for Computational Linguistics, 11 : 100-118.

Sutton R. S.,Barto A. G. .2018 .Reinforcement Learning: An Introduction (2nd ed. , : .

Zhang M.,Wang Y. .2024 .Context-aware argumentation in dialogue systems: A hybrid neurosymbolic framework. Artificial Intelligence Review, 57(3) : 289-313.

Zhao X.,Lee K. .2020 .Data-driven decision-making and customer loyalty in AI-based platforms. Journal of Business Research, 118 : 460-470.

Zhou Q.,Wang L. .2024 .Towards transparent AI: Explainability in conversational recommender systems. Information Processing & Management, 61(1) : 103246.

Zhou R.,Huang X.,Zhao L. .2021 .Intelligent dialogue agents for e-commerce: A survey. ACM Transactions on Information Systems, 39(3) : 1-37.