Orgverse: A Temporal And Multidimensional LLM Framework for Enterprise Data Integration And Decision

Inesh Hettiarachchi ()

Journal of Artificial Intelligence General science (JAIGS) ISSN:3006-4023, 2025, vol. 8, issue 02, 245-261

Abstract: Enterprises generate vast streams of heterogeneous data across transactional, operational, and human resource domains. Traditional business intelligence pipelines remain fragmented and retrospective, often failing to support real-time decision-making. This paper introduces OrgVerse, a temporal and multidimensional LLM framework for enterprise data integration and decision support. OrgVerse models organizational data as a temporally versioned knowledge universe with vectorized entities and relationships, enabling accurate “state-as-of” reasoning. By combining temporal retrieval-augmented generation (RAG), relation-aware embeddings, and agentic role-based orchestration, OrgVerse delivers interactive, multidimensional (3D/4D) reports while enforcing privacy and compliance. We outline its architecture, data representation methods, training strategies, governance design, and evaluation criteria, and argue that OrgVerse offers a novel pathway toward dynamic, explainable, and role-adaptive decision support for enterprises. Building upon the progress of the large language models and enterprise knowledge graphs, OrgVerse harmonizes the structured and unstructured data in one temporal semantic layer that supports descriptive and predictive analytics. Its design takes into account the need for dynamically updating the organizational states, so that the decision-makers can track the changes in the past, project the future outcomes and adapt the strategies in near real-time. Furthermore, the framework includes differential privacy techniques and considers federated learning strategies to ensure compliance with international data protection standards such as GDPR and HIPAA from which trust and adoption can be ensured across industries. We describe its architecture, data representation techniques, training techniques, design strategies for organizational governance, and evaluation metrics; and make the case for OrgVerse as a new scientific direction for dynamic, explainable, and role-adaptive decision supports for enterprises. This piece of research helps create a gap between the old, static nature of business intelligence and the adaptive and LLM- based nature of pertinent knowledge ecosystems where making change in OrgVerse becomes a canal or a foundational model for the upcoming generations of enterprise digital transformations.

Keywords: Temporal Large Language Models (LLMs); Enterprise Data Integration; Retrieval-Augmented Generation (RAG); Multidimensional Knowledge Representation; AI Governance and Compliance (search for similar items in EconPapers)
Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://newjaigs.com/index.php/JAIGS/article/view/419 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:das:njaigs:v:8:y:2025:i:02:p:245-261:id:419

Access Statistics for this article

Journal of Artificial Intelligence General science (JAIGS) ISSN:3006-4023 is currently edited by Justyna Żywiołek

More articles in Journal of Artificial Intelligence General science (JAIGS) ISSN:3006-4023 from Open Knowledge
Bibliographic data for series maintained by Open Knowledge ().