Supercharge Your Innovation With Domain-Expert AI Agents!

Predicted value correction method based on collaborative filtering

A technology of collaborative filtering and correction methods, which is applied in the fields of instruments, calculations, electrical digital data processing, etc., can solve the problems of high algorithm complexity and achieve the effect of small time complexity

Inactive Publication Date: 2013-09-04
EAST CHINA NORMAL UNIV
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But when the user-item rating matrix is ​​extremely sparse, the complexity of the algorithm for correcting unknown ratings is extremely high

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
  • Predicted value correction method based on collaborative filtering
  • Predicted value correction method based on collaborative filtering
  • Predicted value correction method based on collaborative filtering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0044] Assuming that 1, 2, 3, 4, and 5 rating levels are set in the movie recommendation scoring system, and a historical dataset of movie ratings for a user is given, in order to make the most accurate recommendation for a certain user, it is first necessary to predict a certain user’s rating for a certain user. Item-specific ratings, and then make reasonable recommendations based on the predicted ratings. Generally, the predicted ratings are in decimal form. In the recommendation process, in order to show the user the basis for recommending this movie, it is necessary to show the degree of prediction of the user's preference for the movie. Generally speaking, the number of stars is to measure, so it is also necessary to correct the predicted score to an integer corresponding to a certain score level. The proposed correction method is based on the distribution of the predicted ratings of the training data set, specifically, the distribution of the respective predicted ratings...

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 predicted value correction method based on collaborative filtering. The method includes the following steps: a. determining a probability distribution condition of predicted grades of different grade levels in a training set; b. designing and determining a target function when the prediction grades are a certain integer; and c. conducting predicted value correction on the predicted grades to enable the mean absolute error to be minimum. The larger the target function value is, the more accurate the selected integer is. The method is a predicted value correction method based on the probability distribution of the predicted grades in the training set and has the advantages of higher accuracy, smaller time complexity and forceful theoretical basis.

Description

technical field [0001] The present invention relates to the technical field of information recommendation, in particular to a method for correcting predicted values ​​based on collaborative filtering, especially a method for correcting predicted values ​​based on the probability distribution of predicted ratings in a training set. Background technique [0002] As the most successful personalized recommendation technology, collaborative filtering algorithm has been applied in many fields. The predicted value generated by the algorithm is usually a decimal number. In the process of using the training scoring data to train the predictive model, due to the characteristics of the score distribution of the training data set, there will be a certain deviation between the score predicted by the predictive model and the real score. Correcting the predicted rating to a more accurate integer can reduce the difference between the predicted rating and the real rating, and usually the rec...

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): G06F19/00
Inventor 贺樑王伟杰向平李明耀陈国梁杜泽宇
Owner EAST CHINA NORMAL UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com