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

User session recommendation method based on deep neural network

A deep neural network and recommendation method technology, which is applied in the field of user session recommendation, can solve problems such as inapplicable recommendation services, and achieve the effect of improving recommendation accuracy and efficiency

Inactive Publication Date: 2016-07-20
ZHEJIANG UNIV
View PDF1 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, it is not suitable for recommendation services in interactive scenarios

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
  • User session recommendation method based on deep neural network
  • User session recommendation method based on deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] The present invention will be further described below in conjunction with drawings and embodiments.

[0039] Such as figure 1 Shown is a work flow diagram of a method for recommending a user session based on a deep neural network in the present invention, and the steps of the method are as follows:

[0040] Step 1) Collect the log data of the shopping website, classify the log data according to the same session of a user, reconstruct the product pages browsed by all users in the complete session, and generate a training set D according to the ratio of 6:2:2 ', cross-validation set V' and test set T', (according to experience, the proportion of the general training set is greater than the proportion of the test set of the cross-validation set, and the proportion of the general cross-validation set and the test set is the same);

[0041] Step 2) expand the complete sessions of all users, and use the complete sessions to generate multiple incomplete sessions, thereby gene...

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 user session recommendation method based on a deep neural network. The method comprises steps as follows: collecting shopping website log data, reconstructing commodity webpages browsed by user sessions, and generating a training set D', a cross validation set V' and a test set T'; extending all the sessions so as to generate a training set D, a cross validation set V and a test set T; setting neutral network parameters and generating a deep neutral network model; training a neutral network and performing parameter adjustment with an automatic parameter adjusting framework based on the genetic algorithm; calculating the correct rate of the model on the test set T; calculating a path compression ratio. With the adoption of the method, the recommendation accuracy can be significantly improved, the problem about difficulty in neutral network parameter adjustment is solved, the algorithm for automatically adjusting network related configuration is provided on the basis of the genetic algorithm, and the network parameter adjustment efficiency is greatly improved. The method is applicable to online shopping websites to recommend commodities which users may buy to the users while the online shopping websites interact with the users.

Description

technical field [0001] The invention relates to a user session recommendation method, in particular to a user session recommendation method based on a deep neural network. Background technique [0002] To deal with the problem of information overload, people increasingly rely on recommender systems to select the information and items we need. For example, Taobao, JD.com, Kaola.com, Suning.com, Amazon and other shopping sites provide users with a variety of recommendation services in different forms. It recommends items that the user is likely to purchase. [0003] In recent years, collaborative filtering algorithms have been adopted by many practical systems. P.Resnick et al. (1994) studied the collaborative filtering recommendation algorithm earlier, loading the entire user-item data set into memory, and treating it as a whole for rating prediction. The algorithm uses the Pearson correlation coefficient as the similarity weight between two users, and then predicts the ta...

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): G06F17/30G06Q30/06G06N3/02
CPCG06F16/9535G06N3/02G06Q30/0631
Inventor 寿黎但陈珂陈刚胡天磊伍赛俞骋超
Owner ZHEJIANG UNIV
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