基于Fasttext网络的煤矿事故案例文本分类方法对比  

Comparison of text classification methods of coal mine accident cases based on Fasttext network

作　　者：闫琰 杨梦 周法国[1] 葛逸凡 YAN Yan;YANG Meng;ZHOU Fa-guo;GE Yi-fan(School of Mechanical Electronic and Information Engineering,China University of Mining and Technology(Beijing),Beijing 100083,China)

机构地区：[1]中国矿业大学(北京)机电与信息工程学院,北京100083

出　　处：《煤炭工程》2021年第11期186-192,共7页Coal Engineering

基　　金：中央高校基本科研业务费专项资金资助(8000150A082)。

摘　　要：随着大数据时代的到来,煤矿企业积累了大量煤矿数据资源。其中煤矿事故案例信息丰富,包括对事故发生时间、事故发生经过、导致事故的原因以及事故报告等多个方面的分析和总结,但是对这些非结构化文本信息提取很困难,不能有效的获得隐含的语义特征。因此针对煤矿事故案例,对比基于Fasttext网络的文本表示与分类方法,更好的挖掘文本中的语义信息,并准确有效的对案例事故进行类别预测,为后续建立专家知识库、构建应急救援平台提供有力的技术支撑。该文实验的所有代码均已放在GitHub上。With the rapid development of the big data, coal mining enterprises has accumulated a large amount of coal mine data resources. Among them, the coal mine accident case information is abundant, including the analysis and summary of the time, course and reasons of the accidents, and the accident reports. However, it is difficult to extract unstructured data information, and the implicit semantic features cannot be efficiently obtained. Therefore, we focus on coal mine accident cases and compare the text representation and classification methods based on Fasttext network, to better mine the semantic information in the text, and then accurately and effectively predict the category of the case accident. Thus to provide strong technical support for the subsequent establishment of expert knowledge base and construction of emergency rescue platform. All the code for the experiment in this paper has been placed on GitHub.

关 键 词：深度学习 煤矿案例 文本分类 文本表示 

分 类 号：TD-05[矿业工程] TP39[自动化与计算机技术—计算机应用技术]