Method and system for query intent mining

A technology of query intent and intent, applied in the field of information retrieval, can solve the problem of inability to provide accurate intent mining solutions

Inactive Publication Date: 2017-04-19
TSINGHUA UNIV +1
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It can be seen that the THUIR system cannot provide a high-accuracy intent mining solution

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
  • Method and system for query intent mining
  • Method and system for query intent mining
  • Method and system for query intent mining

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0119] Such as Figure 4 Shown is a schematic diagram of intent mining based on numerical query examples in the method for query intention mining of the present invention, where,

[0120] S1000: Get the user's search query;

[0121] For example: cipro for uti 4days.

[0122] (cipro: ciprofloxacin, uti: urinary tract infection)

[0123] S1100: Identify key concepts in search queries;

[0124] Recognized concepts: cipro, uti, day.

[0125] S1200: Identify at least one numeric type including period, frequency, distance, amount, or level from the key concepts;

[0126] Recognized value type: day-> numeric type=period.

[0127] S1300: For each recognized value type, generate at least one value query instance including a value query structure; the value query structure includes key concepts, value types, and instance values; in this embodiment, as a preference, The numeric query structure only includes the key concepts, numeric types, and values ​​of the search query, that is, it is composed of...

Embodiment 2

[0272] When the technical solution of the present invention is integrated into the existing technology, such as Picture 11 As shown,

[0273] S2000: Get search query;

[0274] S2100: Identify key concepts in search queries;

[0275] S2200: Determine whether there are numeric types in the key concepts mentioned above;

[0276] If there is a numeric type, proceed to steps S2500-S2800, and the steps S2500-S2800 are the same as those in the first embodiment. Figure 4 Steps S1300-S1600 in are the same, and will not be repeated here.

[0277] If there is no numeric type:

[0278] S2300: Mining candidate intents related to search queries from search results, Wikipedia, and click data;

[0279] S2400: Sort the candidate intents. Features include intent frequency, co-occurrence frequency, click statistics and edit distance;

[0280] Step S2800 or step S2400 is followed by step S2900 to output the intent list.

Embodiment 3

[0282] When integrating the technical solution of the present invention into retrieval, such as Picture 12 As shown,

[0283] S3000: Get search query;

[0284] S3050: Identify key concepts related to search queries;

[0285] S3100: Extract candidate sets of documents related to key concepts;

[0286] S3600: Give the candidate document a relevance score based on the intent list;

[0287] S3700: Sort candidate documents according to the relevance score;

[0288] Perform steps S3150-S3500 after step S3050 and before step S3600;

[0289] Steps S3150-S3500 and Example 2 and Picture 11 Steps S2200-S2900 are the same, and will not be repeated here.

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 relates to a query intention mining method. The method includes the steps of acquiring a search query, recognizing a key concept in the search query, recognizing numerical value types in the key concept, generating a numerical value query instance for each recognized numerical value type, mining a corresponding candidate intention from a data source, calculating a value range of the corresponding candidate intention through the numerical value query instances, clustering the candidate intentions, and outputting an intention list. The invention further discloses a query intention mining system. The system comprises a search query acquisition module, a key concept recognition module, a numerical value type recognition module, a numerical value query instance generation module, a candidate intention mining module, a calculating module, a clustering module and a display module. According to the technical scheme, the query intention mining method and system has the advantages that search results are subjected to effective indexing, organizing and the like according to query intentions of users, accuracy is improved effectively and time and effort waste caused by users screening intentions that are not theirs is avoided.

Description

Technical field [0001] The invention relates to the field of information retrieval, in particular to a method and system for query intention mining. Background technique [0002] The Internet is a platform for the official release of scientific and technological information, and for individuals to post diaries or blogs. Information retrieval systems (such as search engines) are becoming increasingly important because they can find the information users want from big data sets; however, different users will use the same short and vague query term to find different information (interpretation). As a result, it is difficult for the existing information retrieval system to return sufficient and accurate results. In order to help users quickly and accurately find the information they are interested in, various search results sorting methods based on natural language processing and information retrieval have emerged. [0003] Such as figure 1 As shown, a user interface is displayed, in...

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 Patents(China)
IPC IPC(8): G06F17/30
CPCG06F16/9535
Inventor 夏云庆那森黄耀海赵欢
Owner TSINGHUA UNIV
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