A Digital Classification Method of Rural Public Affairs Governance Based on K-means Clustering Algorithm

The diversity of data sources may lead to differences in data format and data quality, which will affect the accuracy of digital classification. Therefore, a digital classification method of rural public affairs governance based on a K-means clustering algorithm is proposed. The drift detection of the digital information of rural public affairs governance is carried out using the inequality of Hoeffding Bounds. The naive Bayes classifier is used to accurately detect the noise data in the digital information. The gray clustering technology is introduced to divide the digital files into three categories: direct use, optimized use and non-use. This paper proposes a digital rural public affairs governance framework based on service agent technology. It also proposes a K-means clustering algorithm based on an effective index, uses the SDClass algorithm to merge and classify the digital information of rural public affairs governance, and builds a digital classification and processing model of rural public affairs governance based on the K-means clustering algorithm. The experimental results show that the classification time of the design methods is lower than 4[Formula: see text]s, the classification accuracy is higher than 97%, the coverage rate is between 97.5% and 98.8%, and the running time is lower than [Formula: see text][Formula: see text]s.