科研众包项目运行模式与平台类型划分研究

Research on Operation Mode and Platform Classification of Scientific Research Crowdsourcing Projects

[目的/意义]为了有效地推动知识交流与科技创新,探讨新兴的科研众包项目运行模式的构建、科研众包平台类型的划分与特点的解析。[方法/过程]文章从科研众包项目的需求及其支撑要素的视角,一方面,归纳科研众包项目参与主体,设计科研众包流程并解析知识流向,构建知识转移为核心的项目运作模式;另一方面,归纳平台的主要特征作为分类要素,单维度划分类型并进行属性间关联分析,同时,通过K-modes算法进行综合聚类,多维度划分平台类别。[结果/结论]研究结果表明科研众包运行模式中发包方和接包方之间存在知识双向多次转移的现象,依据系统结构和功能角度进行平台类型划分中主要以服务层面平台特征最为明显,自主专建型与非专项型关联紧密,嵌入式拓展型与专项领域型关联紧密;综合聚类划分平台为11个类别时效果最佳。

[Purpose/significance]In order to effectively promote knowledge exchange and scientific and technological innovation,this paper discusses the construction of the operation mode of emerging scientific research crowdsourcing projects,the division of the types and characteristics of scientific research crowdsourcing platforms.[Method/process]This study focuses on the perspective that the demand and supporting elements of scientific research crowdsourcing projects.On the one hand,it summarizes the participants of scientific research crowdsourcing projects,designs the crowdsourcing operation process and analyzes the flow of knowledge transfer,and then,it constructs the project operation mode with knowledge transfer as the core.On the other hand,the main characteristics of scientific research crowdsourcing platform are summarized as classification elements.The platform types are divided from one dimension and the association analysis between attributes is carried out.At the same time,the K-modes algorithm is used for comprehensive clustering to divide the main categories of the platforms from multiple dimensions.[Result/conclusion]The research results show that there is a phenomenon of two-way and multiple transfer of knowledge between the employer and the employee.In the classification of platform types from perspective of system structure and function,the service level platform is the most obvious,and the self-established platform is closely related to non-special platform,and the embedded construction platform is closely related to special field platform.When the comprehensive clustering platform is divided into eleven categories,the effect is the best.