“This is the fourth day of my participation in the Gwen Challenge in November. Check out the details: The Last Gwen Challenge in 2021.”

Knowledge map

This concept was introduced by Google in 2012 as a way to upgrade the traditional keyword based search model to semantic based search. Knowledge graph can be used to better query complex association information, understand user intention from semantic level, and improve search quality.

Knowledge representation

Knowledge representation is a common problem in both cognitive science and artificial intelligence.

  • In cognitive science, knowledge representation is concerned with how humans store and process data.
  • In artificial intelligence, the main goal is to store knowledge that programs can process to achieve human intelligence. The field still doesn’t have a perfect answer, but more recently with deep learning, particularly with recurrent neural networks

In knowledge representation, knowledge graph is a knowledge base in which the data is integrated through the data model or topology of graph structure. Knowledge graphs are often used to store entities that are related to each other.

Application of knowledge graph

  • Google search, Baidu search,
  • The linkedin Economic Graph for the social Facebook space
  • Enterprise information in the field of eye enterprise atlas
  • Machine translation
  • financial
  • medical
  • Small startups will invest in opening up the knowledge graph
  • Recommendation system

Knowledge mapping and machine learning

Related fields of knowledge graph

Knowledge maps cover many areas, such as natural language processing,

  • The database
    • RDF database system
    • Data integration and knowledge fusion
  • Machine learning
    • Knowledge Representation of Knowledge Graph Embedding data
  • Knowledge engineering
    • Knowledge Base construction
    • Rule-based reasoning

Related technologies in the knowledge graph

  • Relational database
  • Construction of domain ontology
  • Natural language processing: Natural language processing and knowledge mapping go hand in hand, as knowledge mapping is driven by the emergence of natural language processing today. RDF emerged long ago because of the limitations of manual extraction of information. So you limit the size of the knowledge base. In turn, knowledge mapping also drives the development of natural language processing
    • Natural Language q&A for knowledge graph
    • Chatbot
  • Semantic Web: Making the entire Internet a common medium for information exchange by adding computer-readable semantics (metadata) to documents on the World Wide Web, such as HTML documents.
  • Data mining: An interdisciplinary branch of computer science that uses an intersection of artificial intelligence, machine learning, statistics, and databases to discover patterns in relatively large data sets
  • Machine learning:Knowledge mapandMachine learningWhat kind of chemical reaction will occur when these two seemingly unrelated things are put together
  • Knowledge representation and reasoning
  • Cognitive computing: it is an important part of artificial intelligence and a computer system that simulates the cognitive process of human brain. Cognitive computing represents an entirely new model of computing that involves a host of technological innovations in information analysis, natural language processing and machine learning, enabling decision makers to uncover extraordinary insights from vast amounts of unstructured data.
  • Information retrieval and extraction
    • Semantic parsing
  • Knowledge extraction: To obtain knowledge from massive data by extracting information (structured data)
  • Knowledge fusion: Through the alignment, association and merging of multiple related knowledge graphs, it can be called an organic whole and provide more comprehensive knowledge

Knowledge graph and knowledge engineering

Knowledge graph is a new development form of knowledge engineering in the era of Web and big data. The core of knowledge engineering

  • Knowledge base:
  • Inference engine
Construction of domain ontology

An explicit and detailed description of a shared conceptual system, formalized for a specific domain

knowledge

Knowledge is acquired from data through information extraction

Knowledge fusion

By aligning, combining and merging multiple relevant knowledge maps, it is called an organic whole to provide more comprehensive knowledge

Basic terminology

  • Ontology: from philosophy, that is, the study of natural things in philosophy. Ontology is a concept description, formal constraints, we can understand together, finance, will design capital, executives, capital, define these concepts and the correlation between these concepts, financial domain ontology.

  • entity
  • Relationship between
  • attribute
  • RDF is based on triples (object,predicate, object), subject,predicate, and naruto. There are many ways to serialize RDF: JSON XML JSON-LD