大数据开源集成的风险识别与管控定则  

Risk Identification and Control Rules of Big Data Open-source Integration

作　　者：蒋洁[1,2] 孙明宇[1,2] 许颖[2] 

机构地区：[1]南京信息工程大学公共管理学院,江苏南京210044 [2]南京信息工程大学中意网络侵权研究所,江苏南京210044

出　　处：《情报理论与实践》2016年第4期41-44,共4页Information Studies:Theory & Application

基　　金：国家社会科学基金青年项目"云数据隐私侵权问题研究"(项目编号:13CFX083);江苏省法学会自选课题"大数据挖掘的侵权风险及对策研究"(项目编号:SFH2014D19);中国博士后科学基金特别资助项目"云端鉴识的障碍与对策研究"(项目编号:2015T80567)的成果;江苏省高校"青蓝工程"中优秀青年骨干教师培养对象计划资助

摘　　要：大数据开源集成在推进零界域信息联动与灵活的集智创新、提升分享型智能参与者的核心竞争力、揭示动态社交爆发轨迹的同时,暴露出自动化数据分析突破匿名算法、类型化量级处理导致差别待遇、分布式幂律评测制约创新思维等严重问题。亟待构筑以透明仓储、脱敏挖掘、无害流转等为核心的风险管控定则模型,实现万物互联中均衡各方利益的良性循环。Big data open-source integration promotes information linkage and innovation by collective wisdom,enhances core competitiveness of sharing participants,and reveals the bursts trajectory in social communication. However,the big data opensource integration exposes risks which include braking unnamed algorithm through automatically data analysis,causing discrimination through typed assessment and forecasts,and restricting creativeness through distributed power laws assessment. It is urgent to achieve a virtuous cycle of balancing parties＇ interests in interconnection of everything by constructing the risk control model with focus on transparent data storage,data mining without sensitive information,and data circulating inoffensively.

关 键 词：大数据 开源集成 风险识别 风险管控 

分 类 号：G203[文化科学—传播学]