Study on the Application of Machine Learning in Government Service: Take Consumer Protection Service as an Example

The receiving of consumers' complaints is one of the most important work in government service. The classification of complaints for different industries is a bottleneck in the service flow chart, affecting service efficiency, which also leads to low utilization of data information after complaints are accepted. This paper uses machine learning techniques to help government departments solve the difficult issues of receiving and classifying consumers' complaints. In this paper, Naïve Bayesian algorithm and SVM algorithm are introduced to replace the manual classification in the process. In addition, this paper uses text analysis technology and LDA topic model to cluster industry issues. The visualization result helps government departments visually monitor the changes in the consumer environment and hot issues in the industry can be captured promptly and accurately.