基于hMETIS与FP-Growth的协同创新社区领域知识发现方法

基于在线社区的协同创新作为一种汲取用户智慧的有效机制,有利于企业发掘其产品创新方向.然而大数据时代下,信息过载及失真为社区中的知识获取带来了诸多挑战,且已有研究主要聚焦于利用价值较低的高频知识的提取.因此,为了充分挖掘用户经验与创新知识,本文基于社区知识间的共现关系构建了知识超图模型,并利用超图划分算法hMETIS对知识超图进行了探索性分析,同时结合关联规则算法FP-Growth进一步研究了知识间的关联关系,以此获取不同领域的价值知识.最后,本文通过一个实例证明了该方法的有效性.本文的研究实现了对社区知识的深度挖掘与分析,提出的领域知识发现方法对企业社区管理和产品创新极具指导意义.

Identifying domain knowledge in collaborative innovation communities: Based on hMETIS and FP-Growth

As an effective mechanism to draw users' intelligence, collaborative innovation in online community is beneficial for enterprises to find innovation directions. However, in the era of big data, information overload and distortion creates many challenges for knowledge acquisition from community, and most relevant researches focus on extracting the high-frequency knowledge with low utilization. Therefore, this article constructs a knowledge hypergraph model based on knowledge co-occurrence relationship to exploit users' experiences and innovation knowledge. And then we utilize the hMETIS algorithm and FP-Growth algorithm to analyze relationships among knowledge, to further achieve the valuable knowledge in different fields. We demonstrate the proposed method by a case study. This paper has certain significance in knowledge finding and data mining, and provides a guide for community management and products innovation.