Conversational AI Interfaces for Business User Self-Service in Master Data Management
DOI:
https://doi.org/10.15680/IJCTECE.2026.0904011Keywords:
conversational AI, natural language understanding, master data management, self-service, retrieval-augmented generation, human-in-the-loop, data governance, chatbot, large language, data stewardshipAbstract
Centralized master data management programs increasingly face a demand-side bottleneck: business users need frequent, small answers and corrections involving customer, product, vendor, and location records, but routing every such request through a data steward or an IT ticket queue is slow and consumes scarce stewardship capacity on low complexity work. Conversational artificial intelligence interfaces, built on modern natural language understanding and retrieval-augmented generation techniques, offer a way for business users to query and request changes to master data directly in natural language, while preserving the governance and approval controls that centralized master data management depends on. This article presents a structured study of eight conversational interaction patterns for master data self-service: natural language query, guided slot-filling update requests, retrieval-augmented grounding against the golden record, human-in-the-loop approval handoff, confidence-based escalation, multi-turn disambiguation for conflicting entities, proactive data quality nudges, and audit-logged conversational transactions. A reference architecture combines these patterns into a single conversational self-service platform spanning intent recognition, grounded retrieval, governed action execution, and full conversational audit logging. An evaluation study measures self-service completion rate, response latency, user satisfaction, and safe resolution rate across three levels of conversational AI maturity, four interaction modalities, and a range of natural language understanding confidence thresholds, presented through data tables and four distinct three-dimensional chart types spanning grouped bar, scatter, surface, and multi-line trend visualization. The study finds that hybrid architectures combining retrieval-augmented generation with human-in-the-loop escalation achieve substantially higher self-service completion rates than either rule-based or purely automated approaches while maintaining governance safety, that response latency and user satisfaction vary meaningfully by interaction modality, and that safe resolution rate depends jointly on natural language understanding confidence threshold and steward oversight coverage rather than on either alone. The article closes with a discussion of limitations and open research directions, including adaptive confidence calibration, multilingual master data self-service, and longitudinal measurement of conversational interface impact on overall data quality
References
1. Bai, Y., Kadavath, S., Kundu, S., Askell, A., Kernion, J., Jones, A., et al. (2022). Constitutional AI: harmlessness from AI feedback. arXiv preprint arXiv:2212.08073.
2. Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., et al. (2021). On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258.
3. Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D., Dhariwal, P., et al. (2020). Language models are few-shot learners. In Advances in Neural Information Processing Systems 33 (NeurIPS 2020).
4. Chen, M., Tworek, J., Jun, H., Yuan, Q., Pinto, H. P. de O., Kaplan, J., et al. (2021). Evaluating large language models trained on code. arXiv preprint arXiv:2107.03374.
5. DAMA International. (2017). DAMA-DMBOK: Data Management Body of Knowledge (2nd ed.). Technics Publications.
6. Devlin, J., Chang, M. W., Lee, K., and Toutanova, K. (2018). BERT: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
7. Gartner. (2024). Market Guide for Conversational AI Platforms. Gartner Research.
8. Grice, H. P. (1975). Logic and conversation. In Cole, P., and Morgan, J. L. (Eds.), Syntax and Semantics, Vol. 3: Speech Acts, pp. 43 to 58. Academic Press.
9. Hohpe, G., and Woolf, B. (2003). Enterprise Integration Patterns: Designing, Building, and Deploying Messaging Solutions. Addison-Wesley.
10. Jurafsky, D., and Martin, J. H. (2020). Speech and Language Processing (3rd ed. draft). Stanford University.
11. Karpukhin, V., Oguz, B., Min, S., Lewis, P., Wu, L., Edunov, S., Chen, D., and Yih, W. (2020). Dense passage retrieval for open-domain question answering. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP).
12. Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., et al. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. In Advances in Neural Information Processing Systems 33 (NeurIPS 2020).
13. Loshin, D. (2009). Master Data Management. Morgan Kaufmann.
14. Nass, C., and Moon, Y. (2000). Machines and mindlessness: social responses to computers. Journal of Social Issues, 56(1), pp. 81 to 103.
15. OpenAI. (2023). GPT-4 Technical Report. arXiv preprint arXiv:2303.08774.
16. Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C., Mishkin, P., et al. (2022). Training language models to follow instructions with human feedback. In Advances in Neural Information Processing Systems 35 (NeurIPS 2022).
17. Rajpurkar, P., Zhang, J., Lopyrev, K., and Liang, P. (2016). SQuAD: 100,000+ questions for machine comprehension of text. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP).
