Entity perception of Two-Step-Matching framework for public opinions
作者机构:Library and Research Center of Computer Systems ScienceUniversity of Shanghai for Science and TechnologyShanghai 200093PR China School of Accountancy and Shanghai Key Laboratory of Financial Information TechnologyShanghai University of Finance and EconomicsShanghai 200433PR China School of HumanitiesShanghai University of Finance and EconomicsShanghai 200433PR China Institute of Sina WRD Big DataShanghai 201204PR China
出 版 物:《Journal of Safety Science and Resilience》 (安全科学与韧性(英文))
年 卷 期:2020年第1卷第1期
页 面:36-43页
核心收录:
学科分类:1004[医学-公共卫生与预防医学(可授医学、理学学位)] 08[工学] 0835[工学-软件工程] 081202[工学-计算机软件与理论] 0812[工学-计算机科学与技术(可授工学、理学学位)]
基 金:This work is partially supported by the National Natural Science Foundation of China(Grant Nos.71901144,71771152,61773248) the Major Program of National Fund of Philosophy and Social Science of China(18ZDA088,20ZDA060) Shanghai Planning Office of Philosophy and Social Science Foundation(Grant No.2019EXW001) Foundation of University of Finance and Economics(Grant No.2017110709) S-Tech internet communication project(Grant Nos.2018PHD005 and 2018TECH003)
主 题:Entity perception BiLSTM-CRF model Jaro-Winkler distance algorithm User comments Public opinions
摘 要:Entity perception of ambiguous user comments is a critical problem of target identification for huge amount of public *** this paper,a Two-Step-Matching method is proposed to identify the precise target entity from multiple entities ***,potential entities are extracted by BiLSTM-CRF model and characteristic words by TF-IDF model from public ***,the first matching is implemented between potential entities and an official business directory by Jaro-Winkler distance ***,in order to find the pre-cise one,an industry-characteristic dictionary is developed into the second matching *** precise entity is identified according to the count of characteristic words matching to industry-characteristic *** addition,associated rate(global indicator)and accuracy rate(sample indicator)are defined for evaluation of matching *** results for three data sets of public opinions about major public health events show that the highest associated rate and accuracy rate arrive at 0.93 and 0.95,averagely enhanced by 32%and 30%above the case of using the first matching process *** framework provides the method to find the true target entity of really wanted expression from public opinions.