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

A Hybrid Recommendation Method Based on Recommendation Probability Fusion

A hybrid recommendation and probability fusion technology, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve problems such as unfused, reduced recommendation system accuracy, and neglect of user evaluation uncertainty

Active Publication Date: 2016-06-29
HEFEI UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing method expresses the user rating as a specific value, ignoring the uncertainty of user evaluation, unable to reflect the real situation of user evaluation products, and reducing the accuracy of the recommendation system
[0005] (2) Fusion of recommendation information
For example, a collaborative filtering recommendation algorithm based on item rating prediction uses the prediction results obtained by the item-based collaborative filtering method as the input of the user-based collaborative filtering method. method, but it does not fuse the results of user-based collaborative filtering and item-based collaborative filtering

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
  • A Hybrid Recommendation Method Based on Recommendation Probability Fusion
  • A Hybrid Recommendation Method Based on Recommendation Probability Fusion
  • A Hybrid Recommendation Method Based on Recommendation Probability Fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] The present invention uses a two-dimensional table to represent the scoring data of commodities, takes any item in the scored set as an unknown number, uses the basic recommendation method to obtain the prediction result of the corresponding item, and uses the neural network to obtain the score of the scored item and the prediction result of the corresponding item. Perform training to obtain the score prediction model SFM. The set of prediction results of unrated items obtained by the basic recommendation method is used to obtain the final predicted value of unrated items by using the score prediction model SFM. Finally, comparisons are made with the underlying algorithms on standard datasets. Such as figure 1 As shown, the method of the embodiment of the present invention includes the following steps:

[0053] Step 1. Use a two-dimensional table T={U, I, f} to represent the scoring data of the product, specifically including:

[0054] As in Table 1, U={U 1 ,...,U ...

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 hybrid recommendation method based on fusion of recommendation probabilities. The method includes the following steps: 1) using a two-dimensional table to represent the rating data of commodities; Obtain the prediction results of the corresponding items, and use the neural network to train the score prediction model SFM by using the score of the rated items and the prediction results of the corresponding items; Use SFM to obtain the final predicted value of unrated items; 4) Arrange all the unrated items of the user in descending order according to the size of each predicted value in the final predicted value set of unrated items to obtain the sorted set of unrated items, and select unrated items The top N items of the sorted set are recommended to the user as the recommendation result. The invention can effectively reflect the real situation of user evaluation and improve the accuracy of personalized recommendation.

Description

technical field [0001] The invention belongs to the field of electronic commerce, in particular to a hybrid recommendation method based on fusion of recommendation probabilities. Background technique [0002] With the rapid development of e-commerce, the phenomenon of information overload is becoming more and more serious. How to meet the individual needs of users based on a massive collection of commodities has become an important issue to improve user experience and user satisfaction. Personalized recommendation system is an important means to meet the individual needs of users. The personalized recommendation system builds a user interest preference model based on the user's individual online browsing data or purchase data, so as to recommend products that meet their unique needs to users. Personalized product recommendation has been widely used in Amazon, JD.com, Taobao and other e-commerce websites, which effectively increases the possibility of users buying and impro...

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): G06Q30/02G06F17/30
Inventor 刘业政姜元春王锦坤孙春华魏婧杜非王佳佳姬建睿何建民凌海峰
Owner HEFEI 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