应急专家发现路径融合模型探究  被引量:1

A Probe into the Emergency Experts Discovery Path Fusion Model

在线阅读下载全文

作  者:叶光辉[1] Ye Guanghui(School of Information Management of Central China Normal University, Wuhan, 430079)

机构地区:[1]华中师范大学信息管理学院,武汉430079

出  处:《情报资料工作》2019年第2期71-75,共5页Information and Documentation Services

基  金:国家自然科学基金项目"基于‘排名分析-特征识别’的应急决策专家发现"(批准号:71740012);湖北省自然科学基金"基于社会化标签挖掘的智慧城市‘印象云’构建模式研究"(编号:2018CFB387);中央高校基本科研业务费项目"基于社会化标签挖掘的城市画像研究"(编号:CCNU18QN040)的研究成果之一

摘  要:文章鉴于目前应急专家发现存在的问题,提出构建应急专家发现路径融合模型。从融合介入时间节点将模型划分为三个子模式:应急专家关联资源整合在评估资源分布态势的基础之上,设计基于资源均衡度的路径加权融合流程;应急专家特征融合引入工程化建模逻辑,构建从特征选择到特征关联,再到特征融合的分析模式;应急专家发现结果融合在不改变原有发现渠道的同时,通过设计分面指标及分面排名融合算法,获取各路径推荐的综合结果。结合典型场景及业务需求构建应急专家发现路径融合模型,可为应急处置提供全面有效的智力支撑。不同融合模式的横向比较还未进行深入探讨与分析。In view of the current problems in emergency experts finding, the author proposes to build path fusion model. The model can be divided into three sub-modes from the time node of fusion intervention as follow: on the basis of assessing the distribution of resources, path-weighted fusion process based on resource balance degree has been designed in resource integration related to emergency experts;Incorporating engineering modeling logic, the analysis model from feature selection and feature correlation to feature fusion has been built in emergency experts feature fusion;While not changing the original path discovery channel, the baseline finding results are merged by designing the facet index and faceted ranking fusion algorithm to obtain the comprehensive recommendations. Combining the typical scenarios and business requirements, the construction of the path fusion model for emergency expert finding can provide comprehensive and effective intellectual support for emergency response. The horizontal comparison of different fusion modes has not been thoroughly discussed and analyzed.

关 键 词:应急决策 专家发现 特征融合 排名分析 资源均衡度 

分 类 号:G252.7[文化科学—图书馆学] D63[政治法律—政治学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象