Abstract: To solve the problems such as the strong dependency of technical expert and the low accuracy of expert recommendations in sea trial process,a supernetwork-based expert recommendation approach was proposed according to the characteristics of technical activities in problem-solving process.Through the correlation between knowledge attributes,a supernetwork model composed of technical expert network,technical objects network and knowledge content network was constructed,and the relevant technical experts were recommended based on Bayesian method in the supernetwork.To evaluate the performance of the proposed approach,a series of comparison experiments were conducted.Experimental results showed that the proposed approach was more effective in finding experts on the problems in sea trial and improve the accuracy of expert recommendation by comparing with the traditional methods,which fully verified the efficiency and feasibility of the proposed approach.

Key words: sea trial, supernetwork model, expert recommendation, technical knowledge

摘要： 针对船舶试航问题解决过程中存在的专家依赖性强,现有专家推荐方法准确度较低,专家查找困难等问题,从技术问题解决过程的技术活动视角,提出一种基于技术知识超网络模型的专家推荐方法,通过计算技术问题解决过程中知识属性之间的关联度,构建由专家子网、对象子网、知识子网组成的技术知识超网络模型,运用超网络中贝叶斯(Bayesian)推理方法计算技术专家与试航问题间的相关度,并推送相关技术专家。以国内某大型船厂的技术知识和试航问题为实验素材,通过实验比较,表明该方法能更有效地进行船舶试航问题的专家查找,并提高专家推荐的精确度,从而验证了该方法的可行性和有效性。

关键词: 船舶试航, 超网络模型, 专家推荐, 技术知识