摘要
随着互联网的迅猛发展,网络上的文本越来越多,对其进行有效的分类,能方便人们快速获取到有用的信息。但在实际应用中,往往只需针对特定领域的文本进行分类,例如,林业文本分类。对于文本分类这一任务,现在有大量的神经网络方面的优秀模型可供使用,但这些模型常常需要耗费大量的时间、资源进行训练,而朴素贝叶斯这个模型虽然简单,但是,其分类效果已经基本满足工程所需。在原始朴素贝叶斯的基础上,引入类别关键词的因素,能够进一步提升分类的效果。
With the rapid development of the Internet,there are more and more texts on the network.Effective classification can facilitate people to quickly obtain useful information.However,in practical application,it is often only necessary to classify the text in specific fields,such as forestry text classification.For the task of text classification,there are a large number of excellent neural network models available,but these models often need to spend a lot of time and resources for training.Although the Naive Bayes model is simple,however,its classification effect has basically met the needs of the project.Based on the original Naive Bayes,the introduction of category keywords can further improve the effect of classification.
作者
郭肇毅
GUO Zhaoyi(School of Electronic Information and Artificial Intelligence,L eshan Normal University,Leshan Sichuan 614000,China)
出处
《乐山师范学院学报》
2022年第8期39-43,共5页
Journal of Leshan Normal University
基金
乐山市科技局项目“基于人工智能技术的竹编技艺推广系统研究与开发”(21GZD030)。
关键词
林业文本分类
朴素贝叶斯
类别关键词
Forestry Text Classification
Naive Bayes
Category Keywords
