基于加权有限状态机的电话号码规范解析  

STANDARDISATION AND ANALYSIS OF PHONE NUMBERS BASED ON WEIGHTED FSM

作　　者：黄明[1,2] 林家骏[1] 方楠 Lin Jiajun;Huang Ming;Fang Nan(College of Information, East China University of Science and Technology, Shanghai 200237, China;Shanghai 104 Research Institute,Shanghai 200032 , China)

机构地区：[1]华东理工大学信息科学与工程学院,上海200237 [2]上海104研究所,上海200032

出　　处：《计算机应用与软件》2016年第6期76-78,121,共4页Computer Applications and Software

摘　　要：针对社会数据处理中,电话号码数据写法多样,难以有效分析利用的问题,提出一种基于竞争性有限状态机的电话号码解析与规范化方法,并提出相应的基于负反馈的训练算法。经过实际应用检验,该规范化方法的处理速度和正确率都能够满足应用要求,有效解决了在存在输入差异性的场景下,对电话号码进行解析与规范化的问题,具有较好的工程实用性。We proposed a competitive FSM-based phone numbers analysis and standardisation method in light of the problem that in social data processing the phone numbers data are written in various section formats and are difficult to analyse and utilise, and presented the corresponding negative feedback-based training algorithm as well. By verification with practical applications, this standardisation approach can meet the application requirements in both processing speed and accuracy, this effectively solves the problem of analysing and standardising phone numbers under the circumstance with input differences, and has preferable project applicability.

关 键 词：有限状态机 电话号码 文本解析 规范化 负反馈训练 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]