Construction and Application of Financial Knowledge Graphs
This article explains how financial institutions can leverage large amounts of structured and unstructured data to build and apply financial knowledge graphs, covering AI key technologies, schema design, data extraction, graph construction, storage solutions, and real-world use cases such as intelligent tagging, recommendation, policy analysis, and executive relationship mining.
The talk introduces the four key AI technologies—perception, understanding, reasoning, and action—and positions knowledge graphs as an integration of these techniques, enabling machines to store, retrieve, and apply knowledge in financial scenarios.
It outlines the product ecosystem of Daguan Data, including a document intelligent processing platform, an RPA solution, and the Yuanhai Knowledge Graph platform, highlighting capabilities such as multi‑graph management, data source integration, graph construction/editing, graph algorithms, RBAC permissions, and advanced knowledge extraction.
Construction Process
Schema Design : Define entity types, attributes, inheritance, and directed relationships based on business requirements, iteratively refined with domain experts.
Knowledge Extraction : Mapping‑based construction for structured data (e.g., corporate filings, financial statements). Extraction‑based construction for unstructured data using NER (rule‑based, machine‑learning, deep‑learning, BERT‑based) and relation extraction (rule‑based, supervised learning, bootstrapping, distant supervision).
Knowledge Fusion : Merge structured and unstructured extracts, resolve semantic and business inconsistencies across sources.
Knowledge Storage : Use graph databases (Neo4j, JanusGraph) for small scale and distributed storage (HBase, ES, Spark, Kafka) for large‑scale graphs.
Knowledge Application : Enable semantic search, reasoning, recommendation, Q&A, real‑time reporting, decision support, policy analysis, and executive investment relationship mining.
Examples include intelligent labeling of financial news, recommendation and matchmaking based on knowledge graphs, policy graph construction for regulatory insight, and mining hidden investment relationships of listed‑company executives.
The speaker, Wang Wenguang, Vice President of Daguan Data, shares his extensive experience in AI, NLP, OCR, and knowledge graph development, and references additional resources and community links.
DataFunTalk
Dedicated to sharing and discussing big data and AI technology applications, aiming to empower a million data scientists. Regularly hosts live tech talks and curates articles on big data, recommendation/search algorithms, advertising algorithms, NLP, intelligent risk control, autonomous driving, and machine learning/deep learning.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.