PulseGraph

Inspiration

The growing complexity of global events - and their immediate impact on financial markets - drove us to build a solution that captures the world’s pulse in real-time. As an analyst at a leading asset management firm, it became clear that traditional data streams couldn’t keep pace with the sheer volume of news and the ever-evolving geopolitical landscape. Inspired by this gap, we set out to create a platform that unifies diverse global events into one dynamic, interactive knowledge graph, empowering users to make faster, more informed decisions.

What it does

PulseGraph collects, processes, and visualizes the GDELT dataset in real-time, identifying connections between events, locations, and entities. By using ArangoDB’s graph capabilities, it presents an interactive dashboard that highlights emerging risks, sentiment shifts, and patterns across world news. The result is a powerful intelligence tool for financial analysts, risk managers, and decision-makers who need quick insights into potential market impacts and investment opportunities - whether that’s detecting early signals of civil unrest, tracking sentiment changes, or highlighting multinational supply chain disruptions.

How we built it

Data Ingestion: We used GDELT’s open dataset to ingest events, their timestamps, locations, involved entities, and sentiment scores. Graph Modeling: Leveraging ArangoDB, we structured these events as nodes (people, organizations, locations) and edges (relationships or interactions), creating a seamless graph representation of the global news cycle. Backend and API: A Python-based backend cleans and normalizes the data, then loads it into ArangoDB, exposing a query interface for graph analytics. User Interface: A web dashboard visualizes connections, highlights clusters of events, and surfaces risk indicators in an intuitive, real-time display.

Challenges we ran into

Data Complexity: GDELT’s breadth and continuous updates required careful data cleaning and handling of diverse formats to maintain consistency. Graph Design: Structuring highly interconnected data in a way that’s both efficient and insightful demanded iterative schema design and performance tuning. Performance at Scale: Handling tens of thousands of nodes and edges in a user-friendly manner required optimization to ensure near-instant responses to complex queries.

Accomplishments that we're proud of

Real-Time Insights: Transforming a massive real-time news feed into actionable intelligence for risk and investment decision-making. Graph-Driven Discovery: Successfully leveraging ArangoDB’s multi-model capabilities to turn raw, unstructured data into a meaningful knowledge graph. Scalability: Achieving fast query response times despite the continuous influx of new data, ensuring our dashboard remains responsive and user-friendly. Broad Applicability: Designing a solution that, while initially aimed at finance, can also support humanitarian response efforts, policy-making, and research.

What we learned

Importance of Graph Modeling: Properly conceptualizing data in terms of nodes, edges, and relationships opens up powerful new ways to analyze information. Real-Time Data Handling: Handling continuous data streams effectively requires robust data pipelines, efficient storage, and scalable processes. Cross-Disciplinary Collaboration: Building an impactful solution involves a blend of data engineering, UI/UX design, and domain expertise - especially in finance, where real-world applicability is critical.

What's next for PulseGraph

Advanced Analytics: We plan to integrate predictive modeling and machine learning to forecast event impacts and sentiment shifts before they escalate. Expanded Data Sources: Incorporating additional open datasets—from social media feeds to structured economic data - will enrich our insights and increase contextual accuracy. Customizable Dashboards: Providing role-based views and interactive filters, so each user - traders, analysts, policymakers - can tailor insights to their specific needs. API Integration: Enabling easy integration with third-party applications, risk management platforms, and BI tools to foster wider adoption.