基于众包模式的分类法映射研究  

Research on Taxonomy Mapping Based on Crowdsourcing Model

作　　者：陈瑞 贾君枝 Chen Rui

机构地区：[1]武汉大学信息管理学院,湖北武汉430072 [2]中国人民大学信息资源管理学院,北京100872

出　　处：《情报理论与实践》2020年第7期137-143,共7页Information Studies:Theory & Application

摘　　要：[目的/意义]为有效实现分类系统之间的互操作,将众包模式引入分类法的映射过程,充分利用用户认知和群体智慧解决计算机难以单独完成的映射任务,提高映射的准确率和覆盖率。[方法/过程]分析国内外分类法映射方法和众包实践,以自动映射的结果为基础,通过设计众包映射框架、众包映射任务、众包映射质量控制方案等构建众包分类法映射模式,并以具体的实验验证该方法的有效性。[结果/结论]众包分类法映射模式有一定的可行性,自动映射算法提供候选映射类目,有效降低映射难度,众包用户在此基础上进行人工干预,有效纠正错误的映射,扩展更多的映射关系,提高了映射类目的数量和质量。[Purpose/significance]In order to effectively realize the interoperation between classification systems,the crowdsourcing model is introduced into the mapping process of taxonomy,which full utilizes user cognition and group wisdom to solve mapping tasks that is difficult for computers to complete alone,so as to improve the accuracy and coverage of the mapping.[Method/process]By analyzing the mapping method of classification methods and crowdsourcing practices at home and abroad,based on the results of automatic mapping,the crowdsourcing taxonomy mapping model is constructed by designing the crowdsourcing mapping framework,the crowdsourcing mapping task,and the crowdsourcing mapping quality control scheme,etc.Finally,the effectiveness of the method is verified by specific experiments.[Result/conclusion]The crowdsourcing taxonomy mapping model is feasible to some extent.The automatic mapping algorithm provides the candidate mapping categories,which effectively reduces the mapping difficulty.On this basis,crowdsourcing users perform manual intervention to effectively correct the wrong mapping and expand more mapping relationships.This approach improves the quantity and quality of the mapping categories.

关 键 词：众包 分类法映射 自动映射 群体智慧 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术] G254.1[自动化与计算机技术—计算机科学与技术]