STIndex: A Context-Aware Multi-Dimensional Spatiotemporal Information Extraction System

TL;DR

STIndex is an end-to-end system that structures unstructured data into a multidimensional spatiotemporal warehouse, enhancing extraction accuracy and providing interactive analysis tools.

Contribution

It introduces a configurable, context-aware extraction system leveraging large language models for structured spatiotemporal data from unstructured content.

Findings

Improves spatiotemporal entity extraction F1 by over 4%.

Integrates multiple modules for validation and correction.

Provides an interactive dashboard for analysis.

Abstract

Extracting structured knowledge from unstructured data still faces practical limitations: entity and event extraction pipelines remain brittle, knowledge graph construction requires costly ontology engineering, and cross-domain generalization is rarely production-ready. In contrast, space and time provide universal contextual anchors that naturally align heterogeneous information and benefit downstream tasks such as retrieval and reasoning. We introduce \textbf{STIndex}, an end-to-end system that structures unstructured content into a multidimensional spatiotemporal data warehouse. Users define domain-specific analysis dimensions with configurable hierarchies, while large language models perform context-aware extraction and grounding. \textbf{STIndex} integrates document-level memory, geocoding correction, and quality validation, and offers an interactive analytics dashboard for visualization, clustering, burst detection, and entity network analysis. In evaluation on a public health benchmark, \textbf{STIndex} improves spatiotemporal entity extraction F1 by 4.37\% (GPT-4o-mini) and 3.60\% (Qwen3-8B). A live demonstration and open-source code are available at https://stindex.ai4wa.com/dashboard.