摘要
为了提高建科标准系统标准分类的效率和准确性,本文分别采用朴素贝叶斯分类方法和多种深度神经网络分类方法,对30多万个中文标准进行分类训练和分类测试,结果显示采用深度神经网络Kmaxcnn模型分类准确率最高达到81.2%,实验结果表明,利用该模型可以对建科标准进行高质量的分类,节约了人力成本,大幅提高了效率。
In order to improve the efficiency and accuracy of standard classification in architecture scientific standard system,this paper respectively adopts Naive Bayesian classification and various Deep Neural Networks classification to carry out classification training and classification tests on more than 300,000 Chinese standards. The results show that the classification accuracy of Kmaxcnn model is as high as 81.2% by Deep Neural Network. The experimental results show that this model can beneficial to the high-quality classification of architecture scientific standard,it is not only saving the labor cost,but also greatly improving the efficiency.
出处
《福建建设科技》
2019年第3期77-79,共3页
Fujian Construction Science & Technology
关键词
文本分类
神经网络
标准系统
人工智能
text categorization
Neural networks
Standard Management System
Artificial intelligence
