Construction and application of knowledge map of CCKS-2017 industry-Previous article

This tutorial mainly includes the following three aspects content:

An overview of the industry knowledge map, including an introduction to the industry Atlas, application and challenges of the industry knowledge map, and life cycle management of the industry knowledge map.

The key technologies of the industry knowledge map include the relevant technologies for each process in the industry knowledge map life cycle, available tools, and best practices and related components in each process.

The industry Knowledge Atlas is applied in practice, taking the application of financial securities industry as an example to demonstrate the whole process of knowledge mapping from knowledge modeling and knowledge extraction to industry application.

The target audience is the same as our public fan group:

Knowledge Atlas Learners, technical staff interested in the application of knowledge maps in the industry.

The knowledge and data management personnel who want to introduce the knowledge map related technology in the application of various industries, especially the needs of the industry knowledge base construction and the top quiz search and so on.

Management decision makers who want to know how the knowledge map is applied in the industry.

It takes roughly 60 minutes to read through this article, but it's worth the time because you'll be able to:

Understand the concepts associated with the industry knowledge map and its existing applications in the industry, and understand the value it brings to industry applications.

Understand the relevant challenges and lifecycles of the knowledge map application in the industry, and understand the basic goals and related components of the lifecycle processes.

Familiarize yourself with the technical aspects of the industry knowledge map application, understand what existing tools can be used and relevant considerations, and best practices for some industry applications.

We assume that the audience reading this tutorial has the following basic knowledge:

RDF: Resource Description Framework

Extension of the OWL:RDF Schema

SPARQL:RDF Query Language

section I introduction to the Industry Knowledge Atlas

"Things Not Strings"

As we all know, the Knowledge Atlas was proposed by Google in 2012 to optimize search results.

After years of development, the knowledge map in many industries of artificial intelligence has a mature application.

According to the coverage of the Knowledge Atlas, it is mainly divided into general knowledge Atlas and Industry knowledge Atlas.

1.1 General Knowledge Atlas

The Knowledge Atlas proposed by Google is a common knowledge atlas, which is for the whole field. The general knowledge Atlas is mainly applied to the Internet-oriented search, recommendation, quiz and other business scenarios. Because it emphasizes the breadth, it is more emphasis on entities, it is difficult to generate a complete global ontology layer of unified management.

General knowledge Atlas Some common items are as follows:

1.2 Industry Knowledge Map

Palantir, known as "the most mysterious technology company in Silicon Valley", is a representative of the industry Knowledge Atlas, which allows customers to analyze a wide range of sensitive data for semantic correlation to prevent fraud, ensure data security, and so on.

The industry knowledge Map has the following characteristics relative to the general knowledge Map:

A knowledge map for a specific domain.

User target objects need to take into account the various levels of people in the industry, different people's corresponding operations and business scenarios, and therefore need a certain depth and completeness.

The industry knowledge map is highly accurate, and is often used to assist with complex analytical applications or decision support.

There are strict and rich data patterns, and the entities in the industry knowledge map often have more attributes and have industry significance.

Industry data features include:

Data sources: Internal data, Internet data, third-party data.

Multiple data types: Contains structured, semi-structured, unstructured data, with more and more of the latter.