FinDKG: Dynamic Knowledge Graphs with Large Language Models for Detecting Global Trends in Financial Markets

TL;DR

This paper introduces FinDKG, a dynamic knowledge graph generated by a fine-tuned large language model, and demonstrates its effectiveness in financial trend detection and thematic investing through a novel graph analysis architecture.

Contribution

It presents a new open-source LLM-based method for generating dynamic knowledge graphs from financial news and a specialized GNN architecture for analyzing them.

Findings

FinDKG outperforms existing models in link prediction tasks.

KGTransformer effectively detects financial market trends.

The approach surpasses traditional thematic ETF performance.

Abstract

Dynamic knowledge graphs (DKGs) are popular structures to express different types of connections between objects over time. They can also serve as an efficient mathematical tool to represent information extracted from complex unstructured data sources, such as text or images. Within financial applications, DKGs could be used to detect trends for strategic thematic investing, based on information obtained from financial news articles. In this work, we explore the properties of large language models (LLMs) as dynamic knowledge graph generators, proposing a novel open-source fine-tuned LLM for this purpose, called the Integrated Contextual Knowledge Graph Generator (ICKG). We use ICKG to produce a novel open-source DKG from a corpus of financial news articles, called FinDKG, and we propose an attention-based GNN architecture for analysing it, called KGTransformer. We test the performance of the proposed model on benchmark datasets and FinDKG, demonstrating superior performance on link prediction tasks. Additionally, we evaluate the performance of the KGTransformer on FinDKG for thematic investing, showing it can outperform existing thematic ETFs.