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

A user model-based microblog text recommendation method and recommendation device

A user model and recommendation method technology, applied in the fields of natural language processing, information retrieval, and data mining, which can solve the problem that the topic model cannot achieve the recommendation effect.

Active Publication Date: 2020-05-19
TIANJIN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The LDA topic model can better reflect the topics that users pay attention to, but this method cannot avoid the inaccurate modeling caused by the limitation of the number of Weibo texts.
Simply using this user topic model in recommendation cannot achieve the best recommendation effect

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 user model-based microblog text recommendation method and recommendation device
  • A user model-based microblog text recommendation method and recommendation device
  • A user model-based microblog text recommendation method and recommendation device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] A microblog text recommendation method based on user model, see figure 1 , the microblog text recommendation method includes the following steps:

[0040] 101: Obtain microblog data, form a microblog document, and preprocess the microblog document;

[0041] For example: taking Sina Weibo as the research object, select a certain Sina Weibo user as the target user of the embodiment of the present invention, and perform content recommendation on it. Using the published microblog content and forwarded microblog content of the target user and his followers as the research scope of the embodiment of the present invention, assuming that the microblog content released and forwarded by the target user and his followers is the content that the target user likes, it can be used as a research Content analyzes the interests and hobbies of target users. Capture the microblog data published and forwarded by the target user and his followers, and form a microblog document for model b...

Embodiment 2

[0050] Combined with specific calculation formulas, examples, figure 2 The scheme in Embodiment 1 is described in detail. The MCRA algorithm is divided into two sub-algorithms: Target User Modeling Algorithm (TUMA) and Text Recommendation Algorithm (Content Recommendation Algorithm, CRA). See the following description for details:

[0051] 201: Obtain experimental data;

[0052] That is, the microblog text content published and reposted by target users and their followers is captured to construct experimental microblog documents. When crawling the experimental data, use the open application programming interface (API) of Sina Weibo to design a crawler program, select target users, and users who have a relationship with the target users, and form corresponding microblog documents for each user. During specific implementation, other software may also be used to capture the experimental data, which is not limited in this embodiment of the present invention.

[0053] 202: data ...

Embodiment 3

[0089] After the algorithm is designed and realized, the evaluation method of the algorithm is designed to measure the performance of the algorithm. Taking precision rate (Precision), recall rate (Recall), F value and average precision rate (Average Precision, AP) as evaluation criteria, design evaluation methods, evaluate the effectiveness and correctness of the designed algorithm, and evaluate Analyze the experimental results.

[0090] The number of experimental topics is set to 150, and the TOP-N recommendation is performed by changing the value of the number N of recommended microblogs to 10, 20, 30, 40, 50, 60, 70, and 80, respectively. At the same time, in order to test the effect of the MCRA algorithm, the present invention uses the commonly used recommendation based on the LDA model and the user modeling method based on TF-IDF used by John Hannon et al. , F value and average accuracy rate are used as evaluation indexes to compare and evaluate these three algorithms. ...

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 present invention discloses a user model based microblogging text recommendation method and a recommendation apparatus thereof. The method comprises: obtaining microblogging data, forming a microblogging document, and performing preprocessing on the mircroblogging document; establishing a target user theme model according to an LDA theme model, and calculating a matching degree between a candidate microblog and the target user theme model; establishing a target user keyword vector model based on a TF-IDF algorithm, and calculating a matching degree between the candidate microblog and the target user keyword vector model; and calculating a matching degree between the candidate microblog and a target user model with reference to the two matching degrees by using a weighted average method, using the matching degree as a score of the candidate microblog, and sorting on the score. The apparatus comprises an obtaining and preprocessing module, a first calculating module, a second calculation module, and a sorting module. According to the present invention, microblogging information in which a target user can be interested can be found and recommended to the target user, and connection among users is enhanced, so as to improve vitality of microblog.

Description

technical field [0001] The invention relates to the fields of data mining, natural language processing and information retrieval, in particular to a user model-based microblog text recommendation method (Microblogging Content Recommendation Algorithm, MCRA) and a recommendation device thereof. Background technique [0002] At present, there are many methods of personalized recommendation for microblog user modeling, which can be roughly summarized into two types from the perspective of focus: microblog user relationship or microblog user posting text content. Analyze the relationship between Weibo users and make personalized recommendations: By analyzing the relationship of Weibo users in the social network, analyzing their position in the community, analyzing their user influence in the community, and ranking the influence of Weibo users, Blog users make user recommendations. Analyze the text content published by Weibo users: process and analyze the Weibo content published...

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): G06F16/30
CPCG06F16/9535G06Q50/01
Inventor 喻梅徐天一王建荣于健缑小路郭佳
Owner TIANJIN 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