A Personalized Task-Based Dialogue System Based on Structured User Attribute Description
A technology of user attributes and dialogue system, applied in the field of human-computer interaction, can solve the problems of lack of human dialogue flexibility, inability to change, lack of interaction, etc., to achieve the effect of improving interpretability and accuracy, easy to collect, and easy to obtain
Active Publication Date: 2022-05-03
BEIJING UNIV OF POSTS & TELECOMM
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] 1. Existing task-based dialogue systems are purely data-driven, and the system can only learn dialogue patterns based on existing dialogue data, which will lead to the generated responses tending to be generic and stereotyped, that is, the responses generated for all users are Consistent, lacks the flexibility of human dialogue
[0004] 2. At present, most of the descriptions for user personalities are in natural language, which contains a lot of noise. When encoding, they often lack interaction with the dialogue content, that is, the user personality encoding is purely static and cannot change with the dialogue content. change
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0019] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.
[0020] Such as figure 2 As shown, the present invention is a personalized task-based dialogue system based on structured user attribute description, which describes the user's personality in the form of structured triples, and then uses dynamic coding closely related to the dialogue context The method captures the user's personality knowledge, and finally uses the end-to-end dialogue retrieval model combined with dynamic personality coding to select the most appropriate and appropriate reply, thereby avoiding generating a large number of general and stereotyped replies, achieving the effect of thousands of people and thousands of faces, and improving the user's dialogue process. immersion in. It is mainly divided into two modules: user personality dynamic coding module and personalized matching module. The user personality dynamic encoding ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention is a personalized task-based dialogue system based on structured user attribute description, which describes the user's personality in the form of structured triples, and then uses a dynamic coding method closely related to the dialogue context to capture user personality knowledge. Finally, use the end-to-end dialogue retrieval model combined with dynamic personality coding to select the most appropriate and appropriate reply, so as to avoid generating a large number of generic and stereotyped answers, achieve the effect of thousands of people, and improve the user's sense of integration in the dialogue process.
Description
technical field [0001] The invention relates to a personalized task-type dialog system based on structured user attribute description, which belongs to the field of human-computer interaction. Background technique [0002] The development of NLP technology has greatly promoted the research of task-based dialogue systems, but the existing task-based dialogue systems often have the following two defects: [0003] 1. Existing task-based dialogue systems are purely data-driven, and the system can only learn dialogue patterns based on existing dialogue data, which will lead to the generated responses tending to be generic and stereotyped, that is, the responses generated for all users are Consistent, lacking the flexibility of human dialogue. [0004] 2. At present, most of the descriptions for user personalities are in natural language, which contains a lot of noise. When encoding, they often lack interaction with the dialogue content, that is, the user personality encoding is ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/332G06F16/335
CPCG06F16/3329G06F16/337
Inventor 徐蔚然何可清吴亚楠严渊蒙徐红刘思宏刘子君
Owner BEIJING UNIV OF POSTS & TELECOMM