AB
AB
AiBoss

All Categories

project

OpenSPG - A knowledge graph engine jointly developed by Ant Financial and OpenKG

OpenSPG is a knowledge graph engine based on the SPG framework, launched by Ant Group in collaboration with the OpenKG community. OpenSPG integrates the structure of LPG and the semantics of RDF, overcoming the challenges of implementing complex semantics in RDF/OWL, and inheriting the structure of LPG...

What is OpenSPG?

OpenSPG is a knowledge graph engine based on the SPG framework, launched by Ant Group in collaboration with the OpenKG community. OpenSPG integrates the structure of LPG and the semantics of RDF, overcoming the challenges of implementing complex semantics in RDF/OWL while inheriting the advantages of LPG's simple structure and compatibility with big data systems. OpenSPG provides explicit semantic representation, logical rule definition, and operator framework capabilities, supporting pluggable adaptable basic engines and algorithm services from various vendors to build custom solutions. OpenSPG uses efficient knowledge transformation to help improve the value and application value of data, making it suitable for various business scenarios such as finance.

Main functions of OpenSPG

  • Semantic modeling (SPG-Schema): Responsible for the schema framework design of attribute graph semantic enhancement, including subject model, evolution model, predicate model, etc.
  • Knowledge Builder (SPG-Builder)It supports the import of structured and unstructured knowledge, is compatible with big data architecture, and provides a knowledge building operator framework to realize the transformation of data into knowledge.
  • Logical rule-based reasoning (SPG-Reasoner)Abstract KGDSL (Knowledge Graph Domain Specific Language) provides a programmable symbolic representation for logical rules, supporting rule-based reasoning and neural/symbolic fusion learning.
  • KNext programmable frameworkIt provides scalable, process-oriented, and user-friendly component-based capabilities, enabling the isolation of the engine from business logic and domain models, and quickly defining graph solutions.
  • CloudextIt supports business systems to connect to the open engine based on the SDK, build business front-ends, and adapt to custom graph storage/graph computing engines and machine learning frameworks.

OpenSPG's technical principles

  • Semantic representation of attribute graphsThe OpenSPG framework creatively combines the structure of LPG (Labeled Property Graph) with the semantics of RDF (Resource Description Framework), providing a graph representation method that is both simple and semantically rich.
  • Compatibility and progression between knowledge levelsOpenSPG supports the construction and continuous iterative evolution of knowledge graphs in the presence of incomplete data, and supports the construction and management of knowledge graphs when the data is incomplete or changing.
  • The integration of big data and AI technology systemsThe OpenSPG framework effectively connects big data and AI technologies, supporting efficient knowledge transformation of massive amounts of data and enhancing the value of data and applications.
  • Development of Domain Knowledge Models and OperatorsNew business scenarios are based on extended domain knowledge models and the development of new operators to quickly build domain models and solutions.
  • Definitions of logical rules and semantic rulesBased on KGDSL, it defines logical rules, enabling machines to understand and process complex business logic, and supports rule-based reasoning and neural/symbolic fusion learning.

OpenSPG project address

Application scenarios of OpenSPG

  • Financial sectorIn the financial sector, OpenSPG is used for risk assessment, credit rating, market analysis, and anti-fraud to help financial institutions conduct precise risk control and management.
  • Corporate Decision-Making and Operations ManagementIt helps companies conduct market trend analysis, supply chain optimization, and customer segmentation to support strategic planning and operational management.
  • Customer service and personalized recommendationsIn the area of customer service, OpenSPG provides automated question answering and precise customer support. In terms of personalized recommendations, it recommends relevant content and products based on user behavior and preferences.
  • Search engines and natural language processingOpenSPG can enhance the semantic understanding capabilities of search engines, improve the relevance of search results, and enhance the understanding and analysis capabilities of language models in the field of natural language processing.
  • Medical and Health Education ResearchIn the healthcare field, OpenSPG assists in clinical decision-making and drug development; in the education field, it supports personalized learning and academic research.
Back