.png){kind=icon}
.png){kind=icon}
Return
Sensitive Data Detection of Social Network Based on Improved Random Forest Algorithm
.png){kind=icon}
PRE
AI
DOI：10.34028/iajit/23/2/4.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The characteristics of social network sensitive data are complex, which leads to the difficulty of detecting social network sensitive data, so to study the sensitive data detection method of social network based on improved Random Forest (RF) algorithm. Simulate login to social network, and capture social network information by means of web crawler and collector. The Topology-Based Hierarchical Trait (TBHT) topology feature logic algorithm optimized by Naive Bayesian (NB) algorithm is used to extract sensitive data features of social networks from social network information. The RF algorithm is improved by adaptive node splitting, and a sensitive data detection model based on the improved RF algorithm is built by combining the characteristics of social network sensitive data. Social network information is input into the model, and relevant detection results are obtained. The experimental results show that the data acquisition mode using web crawler and collector runs stably and has a large amount of data acquisition, and the extracted data features are efficient. The accuracy of the improved RF algorithm in data classification is more than 97.5%. Therefore, this method is a powerful and practical method for detecting sensitive data of social networks.
Keywords:
Random forest algorithm
web crawler
naive bayes
sensitive data
feature extraction
data detection
Journal
I
IF:
1.1
Papers:
45
Citations:
815