Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Lexical item weight query learning method based on recurrent neural network

A technology of cyclic neural network and learning method, which is applied in the field of data mining and search engines, can solve problems such as difficult integration, and achieve automatic and efficient prediction effects

Active Publication Date: 2016-10-26
DALIAN UNIV OF TECH
View PDF2 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But its shortcomings are also obvious, that is, the assumption of independence between terms is implied. Even if many studies try to use the overall information of the query to consider this dependency in a disguised form, they only stay at the level of feature construction. Integrate it organically at the model level

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
  • Lexical item weight query learning method based on recurrent neural network
  • Lexical item weight query learning method based on recurrent neural network
  • Lexical item weight query learning method based on recurrent neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The present invention is described below in conjunction with accompanying drawing and specific embodiment:

[0027] The general idea of ​​a method for learning weights of query terms based on cyclic neural networks in the present invention is: firstly use the optimal weight labeling method based on genetic algorithm to search for the optimal weight of terms, then construct the feature vector of query terms, and then construct the query term Item weight learning model, the structure of the model is as follows figure 2 As shown, the model is finally used to predict the query term weights.

[0028] A method for learning query term weights based on a recurrent neural network, comprising the following steps:

[0029] S1. Searching for optimal term weights: collect publicly labeled data sets, and use the genetic algorithm-based optimal weight labeling method to obtain optimal term weight values. The optimal weight labeling method is as follows:

[0030] A1. Initialization: s...

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

Provided is a lexical item weight query learning method based on a recurrent neural network, comprising the steps of: S1, searching for an optimal lexical item weight; S2, constructing a query lexical item characteristic vector; S3, constructing a query lexical item weight learning model; and S4, utilizing the query lexical item weight learning model to predict a query lexical item weight. The invention creatively provides the lexical item weight query learning method which can convert a query lexical item weight prediction problem into a sequence labeling problem, and automatically and efficiently predict a query lexical item weight; the increased ranges of a main evaluation index MAP for a data set are respectively 16.8% (Robust04) and 11.8% (GOV2), which prove the effectiveness of the lexical item weight query learning method for a query lexical item weight learning task.

Description

technical field [0001] The invention relates to the technical fields of data mining and search engines, in particular to a method for learning weights of query terms based on a cyclic neural network. Background technique [0002] The performance of current information retrieval models or systems is highly dependent on query comprehension. Therefore, query comprehension technology has become an important research direction in the field of contemporary information retrieval, and one of the key issues is the analysis and prediction of the importance of each term in the query. Since the weight of query terms plays a very important role in the correlation score calculation formula of mainstream information retrieval models, assigning appropriate weight values ​​to each query term can greatly improve the accuracy of retrieval results. The weight prediction of query terms is closely related to the understanding and representation of queries, and involves technologies such as seman...

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): G06N3/04G06N3/08G06N3/12G06F17/30
CPCG06F16/951G06N3/084G06N3/126G06N3/045
Inventor 田利云马云龙林鸿飞
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com