User query intention understanding method, system and computer terminal

A technology for querying intentions and users, applied in computing, digital data processing, special data processing applications, etc., can solve problems such as inability to extract semantic information, weak feature expression ability, and ignore contextual relations, etc., to achieve strong semantic feature expression ability Effect

Inactive Publication Date: 2018-06-29
厦门太迪智能科技有限公司
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional methods are usually based on the bag-of-words model. This method has problems of high latitude and high sparsity. The fe

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • User query intention understanding method, system and computer terminal
  • User query intention understanding method, system and computer terminal
  • User query intention understanding method, system and computer terminal

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0048] Example 1

[0049] figure 1 A schematic flowchart of a method for understanding user query intentions provided by an embodiment of the present invention is shown.

[0050] Step S110, taking the seed concept of the predetermined intent domain as a starting point, crawling the concept related to the domain in the knowledge base and the classification label of the concept.

[0051] Mark a small number of preset query sentences in the predetermined intent domain, and use the concepts in the sentences as seed concepts. The seed concept may include at least one concept in a predetermined query sentence of a predetermined intent domain;

[0052] Take the seed concept as a starting point, crawl the concept related to the field in the knowledge base and the classification label of the concept, and use the parent classification label and sub classification label of the classification label, the inclusion relationship of the classification label to the related concept and the relationship...

Example Embodiment

[0077] Example 2

[0078] figure 2 It shows a schematic flowchart of another method for understanding user query intentions according to an embodiment of the present invention. In this embodiment, the intention field of "cooking" is taken as an example to illustrate the method for understanding the user's query intention.

[0079] In step S210, the knowledge base is crawled according to the seed concept of the predetermined intent domain.

[0080] For example, mark a small number of query sentences related to the "cooking" intent, such as: "the best Chinese food", "how to make sweet and sour pork ribs", this query statement includes "Chinese food" and "sweet and sour pork ribs" "Two seed concepts.

[0081] Search the Wikipedia knowledge base and search for "Chinese cuisine" to find the concept of "Chinese cuisine", the corresponding "Chinese cuisine" classification label, and links to the concepts of "cold dish", "soy sauce", "fry", etc.; check " "Sweet and Sour Spare Ribs" can be ...

Example Embodiment

[0129] Example 3

[0130] Figure 5 It shows a schematic structural diagram of a user query intention understanding system provided by an embodiment of the present invention.

[0131] The user query intention understanding system 20 includes a data crawling module 21, a link graph building module 22, a probability calculation module 23, a concept return module 24, and an intention judgment module 25.

[0132] The data crawling module 21 uses the seed concept of the predetermined intent domain as a starting point to crawl the concept related to the domain in the knowledge base and the classification label of the concept.

[0133] The link graph construction module 22 establishes a concept link graph of a predetermined intent field according to the crawled concept and the classification label of the concept.

[0134] The probability calculation module 23 calculates the probability that a concept node in the concept link graph belongs to a predetermined intent domain.

[0135] The concept r...

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

No PUM Login to view more

Abstract

The invention discloses a user query intention understanding method, system and computer terminal. The method comprises the steps that a seed concept of a predetermined intention domain is taken as astarting point, concepts and classification labels related to the domain in a knowledge base are crawled; according to the crawled concepts and classification labels, a concept link graph of the predetermined intention domain is established; the probability of nodes in the concept link graph belonging to the predetermined intention domain is calculated; a user query statement is acquired, the concept link graph is retrieved, if the user query statement covers concept nodes, the concepts are returned; if no, a deep learning model is used to calculate the similarity between the user query statement and each concept node, and the first K concepts most similar to the user query statement are returned; according to a returned result and the probability of the concept nodes belonging to the predetermined intention domain, the probability of the user query statement belonging to the predetermined intention domain is calculated, and compared with a predetermined threshold, it is judged whetheror not the user query statement belongs to the predetermined intention domain. The user query intention understanding method, system and computer terminal solve the problem of the concept feature coverage and semantic expression of the query statement.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a method, system and computer terminal for understanding user query intentions. Background technique [0002] The understanding of user query intent has always been the core issue in search engine technology, and solving this problem can greatly improve user experience. At the same time, with the rise of artificial intelligence, human-computer dialogue technology has become a hot topic, and this technology also needs to rely on accurate understanding of user query intentions. Therefore, the difficult problem of how to accurately understand user query intent has attracted the attention of both academia and industry. [0003] At present, the mainstream method of query intent understanding is the machine learning method. This type of method first establishes the intent classification system, then obtains the training corpus of each classification, trains th...

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
no application Login to view more
IPC IPC(8): G06F17/30
CPCG06F16/332G06F16/35G06F16/36
Inventor 李成华刘丽君叶正
Owner 厦门太迪智能科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap