MLegis: Demonstrating a Property Graph-Based Information System for Legislative Analytics

Analyzing the complexity and interconnections of legislation poses significant challenges for information systems that must manage richly structured, temporally evolving institutional data. Traditional document-centric approaches fail to adequately capture the relational and dynamic nature of regulatory ecosystems. We present MLegis, an information system engineered to model, manage, and analyze legislative corpora through a property graph conceptualization. The system represents laws, articles, governments, and legislatures as interconnected entities enriched with structural, temporal, and linguistic properties. This modeling approach enables expressive path-based querying, impact analysis, and longitudinal analytics over constantly updating legislation. MLegis adopts a layered architecture comprising a graph data layer, a GraphQL-based service layer, and an interactive frontend supporting advanced statistical analysis, readability assessment, and LLM-assisted reporting. We demonstrate the platform on the Italian legislation use case and evaluate it through a user study assessing usability, perceived utility, and learning impact. Results confirm the effectiveness of the proposed architecture in supporting knowledge-intensive analysis tasks. The proposed modeling and architectural approach is generalizable to other regulatory domains, contributing to the engineering of data-intensive legislative information systems. Tool:https://mlegis.eu/. Video:https://gmql.eu/mlegis_demo.html.