Retrieval-Augmented Generation for Intelligent Pension Query Resolution
International Journal of Enhanced Research in Science, Technology & Engineering · 2026 · 15 (03) · 447-452
DOI: 10.55948/ijerste.2026.c76 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Enhanced Research in Science, Technology & Engineering, 2026. Verfasst von Anup Kagalkar.
Abstract (Originalsprache)
Pension-related query resolution systems frequently encounter challenges arising from fragmented data repositories, evolving policy frameworks, and delays in delivering accurate responses. These limitations reduce system reliability and increase dependency on manual intervention. This study presents an intelligent query resolution framework based on Retrieval-Augmented Generation (RAG), aimed at enhancing the accuracy, contextual relevance, and efficiency of pension-related interactions. The proposed approach combines semantic vector-based retrieval with transformer-driven generative modeling to ensure that responses are both context-aware and grounded in verified policy documents. A comprehensive evaluation is conducted using a large-scale, realworld-like pension dataset, capturing diverse user queries and policy conditions. Performance assessment is carried out through accuracy metrics, response latency analysis, and statistical validation techniques. The results indicate a substantial improvement over conventional rule-based systems and standalone language models, particularly in reducing ambiguity and improving response precision. The framework demonstrates strong potential for deployment in large-scale pension administration systems and digital financial advisory platforms.
Zitieren
APA
Kagalkar, A. (2026). Retrieval-Augmented Generation for Intelligent Pension Query Resolution. International Journal of Enhanced Research in Science, Technology & Engineering, 15(03), 447-452. https://doi.org/10.55948/ijerste.2026.c76
Vancouver
Kagalkar A. Retrieval-Augmented Generation for Intelligent Pension Query Resolution. International Journal of Enhanced Research in Science, Technology & Engineering. 2026;15(03):447-452. doi:10.55948/ijerste.2026.c76
BibTeX
@article{Kagalkar2026,
author = {Kagalkar, Anup},
title = {Retrieval-Augmented Generation for Intelligent Pension Query Resolution},
journal = {International Journal of Enhanced Research in Science, Technology & Engineering},
year = {2026},
volume = {15},
number = {03},
pages = {447--452},
doi = {10.55948/ijerste.2026.c76}
}
EUVIMED