Sorting model establishing method, query automatic completion method and corresponding devices

A sorting model and automatic completion technology, applied in the field of computer applications and intelligent search, can solve the problems of low efficiency and high input cost

Active Publication Date: 2020-06-02
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF10 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Obviously, for the user, the input cost is high and the efficiency is low

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
  • Sorting model establishing method, query automatic completion method and corresponding devices
  • Sorting model establishing method, query automatic completion method and corresponding devices
  • Sorting model establishing method, query automatic completion method and corresponding devices

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0097] image 3 The flow chart of the method for establishing a ranking model provided by the embodiment of the present application, such as image 3 As shown, the method may include the following steps:

[0098] In 301, the query prefix entered by the user when selecting a POI from the query completion suggestion, each POI in the query completion suggestion corresponding to the query prefix, and the POI selected by the user in the query completion suggestion are obtained from the POI query log.

[0099] For example, when a user user_A enters characters one by one to form each query prefix, when entering "ba", clicks the POI "Baidu Building A" from the query completion suggestion, then the user ID user_A, query prefix "ba" is obtained. ", each POI in the corresponding query completion suggestion, and the POI "Baidu Building Block A" selected by the user, as a piece of data. In the same way, many pieces of data can be obtained from the POI query logs of large-scale users for ...

Embodiment 2

[0122] Image 6 The flow chart of the query auto-completion method provided by the embodiment of this application, such as Image 6 As shown in , the method may include the following steps:

[0123] In 601, the query prefix currently input by the user is obtained, and a candidate POI corresponding to the query prefix is ​​determined.

[0124] This application is applicable to various forms of input content, which can be Chinese characters, pinyin, initials, etc., but the input query prefix can be regarded as a character string. As the user enters the query prefix, the query prefix currently input by the user is acquired in real time. For example, when a user wants to input "Baidu Building", he may input multiple query prefixes such as "Baidu", "Baidu", and "Baidu Da", and then execute the method provided by this application for each query prefix. That is, when the user inputs "hundred", the currently input query prefix is ​​"hundred", and the method of the present applicati...

Embodiment 3

[0142] Figure 8 The structure diagram of the device for establishing the sorting model provided by the embodiment of the present application, such as Figure 8 As shown in , the device includes: a first acquisition unit 01, a second acquisition unit 02 and a model training unit 03, wherein the main functions of each component unit are as follows:

[0143] The first obtaining unit 01 is responsible for obtaining from the POI query log the query prefix that the user has input when selecting a POI from the query completion suggestion, each POI in the query completion suggestion corresponding to the query prefix, and the query completion suggestion selected by the user POIs.

[0144] The second acquisition unit 02 is responsible for using the POI selected by the user and the POI not selected by the user in the query completion suggestion corresponding to the same query prefix to form positive and negative example pairs.

[0145] The model training unit 03 is responsible for tra...

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 method for establishing a sorting model, a query automatic completion method and a corresponding device, and relates to the technical field of intelligent search. According to the specific implementation scheme, the method comprises the following steps: obtaining query prefixes input when a user selects POIs from query completion suggestions, the POIs in the query completion suggestions corresponding to the query prefixes and the POIs selected by the user in the query completion suggestions from a POI query log; forming a positive example and negative example pair byutilizing the POIs selected by the user and the POIs not selected by the user in the query completion suggestions corresponding to the same query prefix; training by utilizing the query prefix and thepositive example and negative example pair corresponding to the query prefix to obtain a sorting model; wherein the training target is to maximize the difference between the similarity between the vector representation of the query prefix and the vector representation of the corresponding positive POI and the similarity between the vector representation of the query prefix and the vector representation of the corresponding negative POI.

Description

technical field [0001] The present application relates to the field of computer application technology, in particular to a method for establishing a sorting model, a method for automatic query completion, and a corresponding device in the field of intelligent search technology. Background technique [0002] QAC (Query Auto-Completion) has been widely adopted by mainstream general search engines and vertical search engines. For example, in a map application, when a user inputs a query (query) to search for a certain POI (Point of Interest), it starts from the user inputting an incomplete query (in this application, the incomplete query input by the user is referred to as is the query prefix), the search engine can recommend a series of candidate POIs to the user in real time in the candidate list for the user to choose as the completion result of the query (the query recommended in the candidate list is called query completion suggestion in this application) . Once the user...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/332G06F16/335G06F40/126G06N3/04
CPCG06F16/3322G06F16/335G06N3/045G06N3/08G06N3/044G06F16/332G06F40/274G06F16/909G06N3/04G06N3/082G06F16/3347G06F16/338G06F18/22G06F18/214
Inventor 黄际洲王海峰范淼
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
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