基于改进LDA算法的电力用户咨询文本分类算法  

Text classification algorithm of power user consultation based on improved LDA algorithm

作　　者：李竹青 侯本忠 曹培祥 王一蓉 李向阳 LI Zhuqing;HOU Benzhong;CAO Peixiang;WANG Yirong;LI Xiangyang(State Grid Anhui Electric Power Co.,Ltd,Hefei Anhui 230061,China;State Grid Corporation of China,Beijing 100032,China;Big Data Center of State Grid Corporation of China,Beijing 100032,China;Beijing State Grid Accenture Information Technology Co.,LTD,Beijing 100053,China)

机构地区：[1]国网安徽省电力有限公司,安徽合肥230061 [2]国家电网有限公司,北京100032 [3]国家电网有限公司大数据中心,北京100032 [4]北京国网信通埃森哲信息技术有限公司,北京100053

出　　处：《太赫兹科学与电子信息学报》2024年第12期1400-1406,共7页Journal of Terahertz Science and Electronic Information Technology

摘　　要：针对目前情感极性分析中电力咨询短文本的准确性较低的问题,提出一种基于改进潜在狄利克雷分配(LDA)算法的电力用户咨询文本分类算法。在分析电力咨询短文本与情感的关联关系基础上,定义了基于情感词共现袋、主题特殊词以及主题关系词的概念;为提高语义分析的质量,设计了改进LDA算法的电力用户咨询文本分类算法执行流程。实验表明,所提模型表现出优异性能,平均精确度和平均召回率为90.91%和85.03%。所提模型可充分发挥多模型集成优势,有效提升模型性能。In response to the current issue of low accuracy in sentiment polarity analysis of short texts in power consulting,this paper proposes an improved Latent Dirichlet Allocation(LDA)algorithm-based classification algorithm for power user consulting texts.Based on the analysis of the relationship between power consulting short texts and sentiment,concepts such as sentiment word co-occurrence bags,topic-specific words,and topic relationship words are defined.To improve the quality of semantic analysis,an execution process for the improved LDA algorithm for classifying power user consulting texts is designed.Experiments show that the proposed model demonstrates excellent performance,with an average precision of 90.91%and an average recall rate of 85.03%.The proposed model can fully leverage the advantages of multi-model integration,effectively enhancing the model performance.

关 键 词：电力咨询 文本分类 主题分析 卷积神经网络 潜在狄利克雷分配 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]