Searching method and system for identifying user retrieval intention

A search method and intent technology, applied in the field of information retrieval, can solve the problems of high load, unable to give intent results, different part-of-speech distribution, etc., and achieve the effect of easy implementation.

Active Publication Date: 2013-01-16
深圳宜搜天下科技股份有限公司
View PDF7 Cites 37 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Its shortcomings are: firstly, it has high requirements for search engines, requiring search engines to calculate the score of each click in real time, and the online load will be high; secondly, it requires search engine performance and effect to be stable, and search results can basically meet the needs of users , otherwise the recorded click feedback on the results will be too far from the actual needs of the user; again, the click of the user during the search process is changeable, most of the time it is random, see a link on a topic, an advertisement link, You may click to view it. Such click information is actually a kind of noise, which has little to do with the retrieval request, but it will still be recorded.
This method does not depend on the user's search result information, it is a method of pre-processing, but the shortcoming is how to do different analysis according to different types of retrieval intentions, there is no clear method
[0006] Retrieval intent recognition based on post-mortem search relies too much on search results and user responses, and

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
  • Searching method and system for identifying user retrieval intention
  • Searching method and system for identifying user retrieval intention
  • Searching method and system for identifying user retrieval intention

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] Such as figure 1 Shown is a flow chart of a search method for identifying user retrieval intentions provided by an embodiment of the present invention, in the figure:

[0033] S102. Receive a retrieval request from a user;

[0034] S104. Calculating three metrics of the retrieval request, the similarity of intent features, the degree of entity word association, and the similarity of syntax format;

[0035]Preferably, calculating the similarity of the intent features of the retrieval request in this step may be: performing word segmentation processing on the retrieval keywords of the retrieval request to obtain a retrieval feature vector; calculating the similarity between the retrieval feature vector and the intent feature vector of each type of intent. Among them, the methods for calculating the similarity between the retrieval feature vector and the intention feature vector of each type of intention include but are not limited to: classical cosine distance similarity...

Embodiment 2

[0052] Which type of intention the retrieval belongs to can be expressed by the similarity between the retrieval feature vector Q and the intention feature space IM, such as figure 2 Shown is a flow chart of a method for calculating the similarity of intent features provided by a preferred embodiment of the present invention. In the figure:

[0053] S202. After receiving the retrieval request input by the user, perform word segmentation processing on the retrieval keyword to obtain the retrieval feature vector Q (Q 1 , Q 2 , Q 3 ,...Q s ), where Q j The value of represents the schematic feature F j Whether it is in the retrieval condition.

[0054] Specifically, in this step, the NLP word segmentation technology can be used to perform word segmentation processing on the search keywords. After the word segmentation, a series of vocabulary T (T1, T2, T3, ... Ts) is obtained, and the series of vocabulary is converted into a feature vector Q ( Q1, Q2, Q3, ... Qn). in

[0...

Embodiment 3

[0084] The importance and level of words contained in the retrieval conditions input by the user are different. Relatively speaking, how to distinguish more important words, entity words are more important. If the search keywords contain entity words, the degree of relevance between the entity words and the intent must be calculated. Such as image 3 Shown is a flow chart of a method for calculating entity word relevance provided by a preferred embodiment of the present invention, in the figure:

[0085] S302. Convert the entity word E contained in the retrieval keyword into a vector Ep (Ep1, Ep2, Ep3, ... Epq) for the resource channel P, wherein:

[0086] Ep j = 1 E ∈ P j 0 E ...

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 searching method and a system for identifying user retrieval intention and is applied to the field of information retrieval. The method comprises the following steps: receiving a retrieval request of the user; calculating three measurements, namely the intention characteristic similarity of the retrieval request, entity word association and syntax pattern similarity; determining the user retrieval intention of the three calculated measurements; and searching according to the determined user retrieval intention and outputting a searching result. According to the embodiment of the invention, the similarity between the retrieval words and intention characteristic library is considered, the special function of the entity word and an integral retrieval syntax structure are considered, the retrieval keywords are locally and integrally subjected to intention identification, the information support is provided for a search engine as much as possible, and the method does not completely depend on the result information of the on-line search engine and is easy to realize.

Description

technical field [0001] The invention relates to the field of information retrieval, in particular to a search method and system based on identifying user retrieval intentions. Background technique [0002] The emergence of search engines has given users tools to find information from massive amounts of data. But not every user understands the principles of search engines, so users generally organize their own search keywords to search when using search engines, and change the search keywords to regain search results when the results are not satisfactory. How to enable users to input less and use search engines to obtain the information they need faster is a very important task - how to mine and identify users' potential retrieval intentions according to the retrieval requests entered by users. Once the search engine can grasp the user's search intention, it can use less resources to meet the greater needs of users. [0003] So far, the methods for identifying user retrieva...

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
Inventor 车天文雷大伟石志伟周步恋杨振东王更生王喜民何宏靖徐忆苏
Owner 深圳宜搜天下科技股份有限公司
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