基于改进果蝇优化算法的投诉举报文本分类  

Complaints and reports text classification based on improved fruit fly optimization algorithm

作　　者：范青武 陈光[1,2,3] 杨凯 FAN Qing-wu;CHEN Guang;YANG Kai(Faculty of Information Technology,Beijing University of Technology,Beijing 100124,China;Engineering Research Center of Digital Community of Ministry of Education,Beijing University of Technology,Beijing 100124,China;Beijing Key Laboratory of Urban Rail Transit,Beijing University of Technology,Beijing 100124,China)

机构地区：[1]北京工业大学信息学部,北京100124 [2]北京工业大学教育部数字社区工程研究中心,北京100124 [3]北京工业大学北京城市轨道交通重点实验室,北京100124

出　　处：《计算机工程与设计》2022年第4期1103-1110,共8页Computer Engineering and Design

基　　金：水体污染控制与治理科技重大专项基金项目(2018ZX07111005);国家自然科学基金项目(61873007、61890935)。

摘　　要：为解决投诉举报文本分类困难这一问题,提出一种基于改进果蝇优化算法的文本分类方法。针对果蝇优化算法存在的搜索半径相对固定、种群多样性低等问题,对算法进行改进;采用支持向量机建立文本分类模型,利用改进后的果蝇优化算法对支持向量机的参数进行动态寻优,以此提高模型的分类精度。实验结果表明,该文本分类方法的准确率和召回率相比于文中其它几种对比方法而言均是最高的,验证了其在投诉文本分类问题上具有较高的准确性。To solve the problem of difficulty in the classification of complaints and reports text,a text classification method based on the improved fruit fly optimization algorithm was proposed.The fruit fly optimization algorithm was modified to solve fixed search radius problems and low population diversity.The support vector machine was used to establish the text classification model.The improved fruit fly optimization algorithm was applied to dynamically optimize the support vector machine’s parameters to increase the classification accuracy of the model.Experimental results show that the accuracy and recall rate of the proposed text classification method are the highest compared with other comparison methods,which verifies its high accuracy in the classification of complaints and reports text.

关 键 词：投诉举报 果蝇优化算法 支持向量机 参数寻优 文本分类 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]