Method and device for controlling man-machine conversation based on fusion of knowledge graph and memory graph

A technology of knowledge graph and man-machine dialogue, which is applied in the field of man-machine dialogue control methods and devices, can solve problems such as limited user comprehension ability and inability to meet user needs, and achieve the effect of improving accuracy

CN106599179AActive Publication Date: 2017-04-26EMOTIBOT TECH LTD
5 Cites 5 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2017-04-26

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention provides a method and a device for controlling man-machine conversation based on fusion of a knowledge graph and a memory graph. A private memory graph is added on the basis of a common knowledge graph, so that memory serves as the private knowledge graph, the contents, such as basic information and personal preference, of a user can be stored, and the contents can also be verified and classified. The personal information of the user is memorized, and the privatized and customized private memory graph is serially connected with the common knowledge graph, so that the user is more deeply understood by the system; meanwhile, the personal information is greatly verified, so that the accuracy of memory is greatly improved, and the better service can be provided for the user.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to an artificial intelligence dialogue system, in particular to a method and device for man-machine dialogue control that integrates knowledge graphs and memory graphs. Background technique

[0002] In the existing commercial knowledge graph, the storage content of the graph is mainly recognized knowledge or common sense by the public, and the application method is mainly in the search engine. This kind of knowledge graph is mainly a public knowledge graph.

[0003] In dialogue systems such as chatbots, only responding to users based on public knowledge graphs cannot provide a deeper understanding of users. Contents of the invention

[0004] The purpose of the present invention is to provide a human-computer dialogue control method and device that integrates knowledge graphs and memory graphs, aiming at solving the problem that existing artificial intelligence dialog systems only respond to users based on public knowledge graphs...

Examples

specific Embodiment 1

[0040] Such as figure 1 with figure 2 As shown, the embodiment of the present invention provides a man-machine dialogue control method that integrates knowledge graph and memory graph, including:

[0041] Step S101, receiving information input by a user. The embodiment of the present invention does not limit the information input by the user, preferably, it may be text type information.

[0042] Step S102, obtaining an information entity from the information input by the user; the information entity is knowledge information or personal information. The knowledge information in the embodiment of the present invention includes knowledge and common sense. The personal information in the embodiment of the present invention includes personal basic information and personalized information such as personal preferences.

[0043] Step S103, identifying the information entity.

[0044] Step S104, when the recognition result is knowledge information, verify the knowledge informatio...

specific Embodiment 2

[0052] Such as image 3 As shown, the embodiment of the present invention provides a human-computer dialogue control device that integrates knowledge graphs and memory graphs, including:

[0053] The receiving module 201 is configured to receive information input by a user. The embodiment of the present invention does not limit the information input by the user, preferably, it may be text type information.

[0054] The acquiring module 202 is configured to acquire an information entity from information input by a user; the information entity is knowledge information or personal information. The knowledge information in the embodiment of the present invention includes knowledge and common sense. The personal information in the embodiment of the present invention includes personal basic information and personalized information such as personal preferences.

[0055] The identifying module 203 is configured to identify information entities.

[0056] The verification module 204...