Robustness sorting learning method based on multi-objective particle swarm optimization and application thereof

A multi-objective particle swarm and ranking learning technology, which is applied in the fields of information retrieval and machine learning, can solve problems such as unstable sorting systems, poor performance, and unstable sorting results

Active Publication Date: 2019-07-23
JINGGANGSHAN UNIVERSITY
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

A sorting model with poor robustness will lead to the instability of the sorting system, that is, some queries perform well, while others perform poorly, resulting in unstable sorting results presented to users, and it is difficult to satisfy as much as possible The information needs of different users make it difficult to provide users with a satisfactory experience

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
  • Robustness sorting learning method based on multi-objective particle swarm optimization and application thereof
  • Robustness sorting learning method based on multi-objective particle swarm optimization and application thereof
  • Robustness sorting learning method based on multi-objective particle swarm optimization and application thereof

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0178] Such as Figure 5 As shown, it is assumed that there is a standard sorting learning dataset L2Rdataset, expressed as a set: L2Rdataset={(i , d ij >,y ij )|q i ∈Q,d ij ∈D i ,y ij ∈ Y i ,1≤i≤|Q|,1≤j≤|D i |}, where q i Represents the i-th query, Q represents the limited query set in the ranking learning dataset, |Q| represents the total number of queries in the query set, d ij Represents a collection of documents D i The jth document in D, D i Indicates that it is associated with the query q i collection of documents, |D i |Represents a collection of documents D i The total number of documents in y ij Represents a relevance annotation, reflecting the query q i and document d ij The degree of correlation between, the value can be some level value, such as {1, 2, 3, 4, 5}, Y i ={y i1 ,y i2 ,...,y i|Di|} represents the tag set associated with the query. query-document pair i , d ij > Described by the M-dimensional sorting feature f, which can be expressed...

specific Embodiment 2

[0180] Such as Figure 6 As shown, a robust ranking learning method based on multi-objective particle swarm optimization can be applied to the actual demand sorting application scenarios such as information retrieval, search engine, recommendation system and question answering system. Here, the robust ranking learning method based on multi-objective particle swarm optimization is applied to search engines such as Baidu, Google, Bing, Sogou and Yahoo, as an application example. The ranking model trained by this method is embedded in the ranking system of the search engine, and the ranking model is used to predict the ranking results of the web pages of the query words that users need to search, so as to improve the overall user satisfaction and enhance the user experience.

[0181] The implementation process of the robust ranking learning method based on multi-objective particle swarm optimization applied to the search engine is as follows: Figure 6 As shown, its implementat...

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 robustness sorting learning method based on multi-objective particle swarm optimization and an application thereof, and the method comprises the following steps of 1, designing an effective deviation function and a robustness variance function of a sorting model based on a deviation-variance equalization theory, and constructing two optimization performance indexes of sorting learning; step 2, on the basis of a multi-objective particle swarm optimization algorithm framework, iteratively optimizing the two objectives of the effectiveness deviation function and the robustness variance function of the sorting model on the sorting learning data set to train the sorting model so as to generate a sorting model archiving solution set; and step 3, selecting a Pareto optimal sorting model with a maximum net flow sorting value from the sorting model filing solution set generated in the previous step as a trained final sorting model based on the idea of a preference sequence structure assessment method PROMEHEE II in a multi-attribute decision theory. Compared with the prior art, the method has the advantages of improving the overall user satisfaction, enhancing theuser experience and the like.

Description

technical field [0001] The invention relates to the fields of information retrieval and machine learning, in particular to a robust sorting learning method based on multi-objective particle swarm optimization and its application. Background technique [0002] Sorting learning is to use machine learning technology to automatically train a ranking model to solve the ranking problem. It is a hot issue in the field of information retrieval and machine learning, and has broad application prospects in information retrieval, search engines, recommendation systems and question answering systems. [0003] Due to the dynamics of the Web and the diversity of users' information needs, the performance of some Web search queries may vary greatly under different ranking models, and may suffer a significant loss, thereby degrading the user experience. A robust retrieval system should ensure that the user experience is not compromised by poorly performing queries. Therefore, in order to im...

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): G06N20/00G06N3/00G06F16/9535
CPCG06N20/00G06N3/006G06F16/9535Y02D10/00
Inventor 李金忠夏洁武曾劲涛彭蕾
Owner JINGGANGSHAN UNIVERSITY
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