Key-element-based matrix decomposition and fine tuning method

A matrix decomposition and element technology, applied in the field of matrix decomposition, can solve the problems of key influencing factors that do not take into account the decomposition effect of existing scoring data, etc.

Inactive Publication Date: 2016-09-28
BEIJING JIAOTONG UNIV
View PDF0 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, in the process of learning and optimizing the low-dimensional latent semantic matrix, the traditional algorithm comprehensively considers the influence of the comprehensive score of the entire matrix on the decomposition result in the calculation process, and does not take into account the key influencing factors of the existing score data on the decomposition effect, because in In the initial stage of matrix decomposition, the available elements are only the initial scoring data, so these data should be considered in the decomposition process

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
  • Key-element-based matrix decomposition and fine tuning method
  • Key-element-based matrix decomposition and fine tuning method
  • Key-element-based matrix decomposition and fine tuning method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] The purpose of the present invention is to solve the problems existing in the matrix decomposition algorithm and provide a fast decomposition algorithm based on key elements, and fine-tune the prediction matrix to make the decomposition result more accurate. The present invention decomposes the user-commodity sparse large matrix collected from the data into two matrices in the low-dimensional implicit space, and the initial matrix and the decomposed two matrices meet the minimum difference between the dot products. In the calculation process, the objective function of minimizing this difference is to optimize the two matrices through the gradient descent method, and finally point-multiply the two matrices to obtain the prediction matrix.

[0045] This embodiment provides a processing flow of a matrix decomposition and fine-tuning method based on key elements such as figure 1 As shown, the following processing steps are included:

[0046] Step S110 , collect the user's ...

Embodiment 2

[0072] In this embodiment, the description of the performance of the algorithm is carried out by comparing specific examples:

[0073] 1. Data preparation

[0074] We selected Epioions, a standard scoring data set commonly used in recommendation systems. After preprocessing the data set, we extracted 100,000 pieces of data for simulation experiments. The sparsity of the matrix composed of the experimental data set was 1.5%. The scoring data in the matrix is ​​distributed between 1-5, so we also set the predicted value of the experimental results as a value between 1 and 5.

[0075] Second, form the training set and test set

[0076] Simulation experiments are carried out by means of cross-validation. 90% of the data in the data set is extracted as the training set, and the remaining 10% of the data is used as the verification set.

[0077] Three, the experimental process

[0078] First initialize and randomly generate low-dimensional matrices P and Q, and set the dimension...

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 provides a key-element-based matrix decomposition and fine tuning method. The method comprises: scoring information of a commodity by a user is collected and a sparse scoring matrix R is formed based on the scoring information; according to the sparse scoring matrix R, two low-dimensional matrixes P and Q are initialized, multiplying of the low-dimensional matrixes P and Q is carried out to obtain a predicted matrix R^, and an objective function based on minimization of a difference value between the sparse scoring matrix R and the predicted matrix R^; and the objective function is optimized by an iterative process to obtain the low-dimensional matrixes P and Q and the predicted matrix R^, and fine tuning and updating are carried out on the predicted matrix R^, thereby obtaining a final predicted matrix. According to the invention, in order to overcome defects in the existing matrix decomposition algorithm, a quick matrix decomposition and fine tuning algorithm is put forward based on improvement of the algorithm. With the method, the decomposition efficiency and the decomposition speed of the matrix are increased and the recommendation accuracy is enhanced.

Description

technical field [0001] The invention relates to the technical field of matrix decomposition, in particular to a matrix decomposition and fine-tuning method based on key elements. Background technique [0002] In today's era, the Internet is developing rapidly and has become an important part of people's lives and an important carrier for the spread and development of human civilization. In this era, both information producers and information consumers have encountered great challenges. For information consumers, it is very difficult to find the information they are interested in from a large amount of information; for information producers, it is also difficult to make the information they produce stand out and attract the attention of users. The personalized recommendation system is an important tool to solve this contradiction. The recommendation system provides different services for different users, bringing about a transformation from "people looking for information" t...

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): G06Q10/04G06Q30/02G06F17/30
CPCG06Q10/04G06F16/9535G06Q30/0271
Inventor 刘云张致远熊菲
Owner BEIJING JIAOTONG UNIV
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