UCGIS Symposium 2026

Half-day, hands-on workshop on AI-powered social media analytics using multimodal embeddings and knowledge graphs in Neo4j. Participants will use a curated Twitter/X-style dataset to build AI agents that utilize vector search and graph structures for in-depth analysis. Prior tutorial context: https://github.com/xbwei/data-analysis-with-generative-ai

This workshop requires registration - click here to register

Abstract

Building on the instructor’s earlier workshops on generative AI in social analytics (see: https://github.com/xbwei/data-analysis-with-generative-ai), this hands-on session introduces participants to advanced techniques in multimodal embeddings and Agent-Based Knowledge Graphs. While previous versions focused on flat storage (e.g., MongoDB), this workshop emphasizes the need to model social data as a connected network that AI Agents can use for reasoning and analysis.

To maximize learning time, the workshop uses curated, pre-packaged datasets. This enables participants to bypass API restrictions and focus directly on analytics. Participants will learn how to organize social data into a Neo4j knowledge graph, representing users, posts, and interactions as interconnected nodes instead of isolated documents.

The core of the workshop examines the intersection of Agentic AI and Network Science. Attendees will create high-dimensional vector embeddings for text and images using modern embedding models and store them directly within graph nodes. This allows the development of Graph-Augmented AI Agents—systems capable of performing semantic vector searches and navigating the graph network to retrieve context. By combining these technologies, researchers can build agents that synthesize insights from connected nodes, enabling more nuanced detection of communities and misinformation than traditional methods.

By the end of the session, attendees will gain practical skills in:

• Graph Database Foundations: Setting up a Neo4j database and importing social data structures (Nodes & Edges).
• Multimodal Embeddings: Creating and saving vector embeddings for text and images within the graph to facilitate semantic search.

• Agent-Based Reasoning: Developing agent workflows that leverage vector retrieval and graph traversal to enable evidence-based analysis.

This workshop introduces AgentLoom, a dual-helix governance framework for reliable agentic AI in geospatial programming. Participants will leverage Persistent Knowledge and Enforceable Behavioral Constraints to stabilize LLM outputs, ensuring scientifically rigorous and reproducible software development across complex geospatial workflows.

Prerequisite Knowledge & Materials:
● Intermediate understanding of programming
● Laptop with a modern web browser
● Ideally access to an LLM API key (details to obtain these will be provided) We will also provide a free-tier option

This workshop requires registration - click here to register

Abstract
The transition from passive, chat-based interfaces to autonomous Agentic AI has revealed a critical reliability gap in scientific software production (e.g. application development or programming-based data analysis). While Large Language Models (LLMs) demonstrate remarkable proficiency in generating localized code snippets, they consistently struggle with the structural requirements of software development. These models frequently fail to maintain architectural coherence across long-context development cycles, lack the "memory" to preserve scientific constraints across multiple sessions, and exhibit stochastic variability that undermines the reproducibility of complex geospatial code.

This workshop introduces a dual-helix governance framework designed to move beyond "prompt engineering" toward executable protocols for reliable agentic AI in geospatial contexts. The framework stabilizes agentic execution by decoupling the LLM’s reasoning capabilities from its volatile internal state through two orthogonal axes: Persistent Knowledge Externalization (auditable domain-specific memory) and Enforceable Behavioral Constraints (machine-executable protocols rather than suggestive instructions). The framework is implemented as an open-source software AgentLoom1 that implements a 3-track architecture (Knowledge, Behavior, and Skills). This serves as the structural foundation for building a project-specific Knowledge Graph that functions as a persistent, version-controlled and auditable repository of domain facts, architectural protocols, and validated workflows, ensuring that the agent’s reasoning remains grounded and scientifically rigorous across extended development and interaction cycles.

Participants will explore this framework and its open-source implementation that addresses five fundamental LLM limitations:
1. Long-context Fragmentation: Managing codebases that exceed the effective attention window of modern transformers.
2. Cross-session Forgetting: Maintaining knowledge over multi-day development cycles.
3. Output Stochasticity: Standardizing architectural patterns to ensure predictable, reproducible outputs.
4. Instruction Following Failure: Enforcing strict protocols (e.g., geospatial standards, or accessibility features).
5. Adaptation Rigidity: Facilitating the transparent evolution of a domain knowledge graph without the need for expensive model fine-tuning.

Learning Outcomes:
● Set up AgentLoom’s Knowledge/Behavior/Skills tracks for a geospatial project.
● Identify common reliability failures in LLM-assisted geospatial coding
● Externalize key domain knowledge and project rules into auditable, version-controlled artifacts
● Apply enforceable protocols (checks/tests/constraints) to make agentic outputs more consistent and scientifically valid

This session introduces the foundations of AI within Esri’s ArcGIS platform. It explores how machine learning, deep learning, AI assistants, and large language models integrate with ArcGIS Pro , ArcGIS Online and ArcGIS apps. We will clarify key terminology, examine practical GIS-focused AI workflows, and distinguish between hype and classroom-ready capabilities. Attendees will see examples such as feature extraction, predictive modeling, and automated spatial analysis, along with guidance on positioning AI concepts within existing GIS curricula.

This workshop requires registration - click here to register