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

A Method for Academic Paper Recommendation Based on Deeply Aligned Matrix Factorization Model

A matrix decomposition and paper technology, applied in the field of academic paper recommendation, can solve the problems of paper cold start and data sparseness

Active Publication Date: 2020-07-10
NANJING UNIV
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] Purpose of the invention: In order to overcome the problem of "sparse user-paper interaction data" and paper cold start in the existing collaborative filtering recommendation method, the present invention uses a new type of hybrid recommendation algorithm, adding the text content of the paper to the recommendation based on collaborative filtering In the algorithm, the information of two data sources of "user-paper interaction" and "text content of papers" are used at the same time to recommend academic papers of interest to users

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 Method for Academic Paper Recommendation Based on Deeply Aligned Matrix Factorization Model
  • A Method for Academic Paper Recommendation Based on Deeply Aligned Matrix Factorization Model
  • A Method for Academic Paper Recommendation Based on Deeply Aligned Matrix Factorization Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0114] All the steps in this embodiment run on the ubuntu14.04 platform, use the python language, tensorflow1.2GPU version library, and conduct experiments on two data sets CiteULike-a and CiteULike-t in the field of academic paper recommendation.

[0115] The experimental configuration is: operating system Ubuntu 14.04, memory 32G, 4 TitansX graphics cards.

[0116] The experimental data is prepared as follows: The present invention uses two datasets, CiteULike-a and CiteULike-t, respectively organized by two research groups. Their statistics are shown in Table 1. Both datasets are compiled from the academic social networking site CiteULike. The site allows each researcher user to create their own personal online library of papers they are interested in, and each paper contains text information such as its title and abstract.

[0117] When constructing the "paper-user interaction" matrix, CiteULike-a only retains users who have collected more than 10 papers, while CiteULike...

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 deep alignment matrix decomposition model-based academic paper recommendation method. The method comprises the following steps of: respectively mapping a user and a paper andthe paper and a word into low-dimensional feature spaces with a same dimensionality through two nonlinear multilayer sensors; building a bridge between the two nonlinear multilayer sensors through maximizing a similarity between two low-dimensional expression vectors of the same paper; and finally transferring information through the paper information bridge and training the two sensors in turn.Through the method, when the sensor of the user-paper part is trained, information of the paper-word part can be utilized, and when the sensor of the paper-word part is trained, information of the user-paper part can be utilized, so that user-paper collection records and paper content text information can be used at the same time so as to jointly make contribution to user paper recommendation.

Description

technical field [0001] The invention relates to a method for recommending academic papers based on a deep alignment matrix decomposition model. Background technique [0002] At present, with the development of the Internet, there are a large number of academic papers on the Internet, causing researchers to face a serious problem of information excess. It is estimated that as of 2014, there are already tens of billions of academic papers online, and the number is still growing at more than 6,000 every day. How to help researchers and users quickly find articles they may be interested in from such a large number of papers has become a problem that people pay attention to. [0003] Most of the current related work adopts the method based on keyword retrieval, and treats academic papers as ordinary web pages. However, these methods both ignore the structural features of the paper itself, and do not model users' personalization. In recent years, with the rise of social network...

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): G06F16/9536G06F40/30
CPCG06F40/30
Inventor 戴新宇戴瑾黄书剑张建兵尹存燕陈家骏
Owner NANJING 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