在高科技项目管理领域，项目评审已成为推动项目发展的主要工作方式。然而，人工选择评审专家易受主观因素的影响，存在过程繁琐、效率低、易形成“小圈子”等缺点。 本文从项目管理的角度出发，综合考虑用户选择偏好、项目与专家的知识匹配度、项目间的相似性以及专家的历史表现等因素，融合了基于内容特征、潜在主题特征、协同过滤和用户选择偏好等评审专家推荐算法，构建了一种改进的组合策略的评审专家推荐算法，并通过设计实验，利用航天项目进行了实验研究，结果分析表明本文的推荐算法是可行且实用的。

In the field of high-tech project management, project review has become the main way to promote project development. However, manual selection of review experts is easy to be affected by subjective factors, which has the disadvantages of tedious process, low efficiency and easy to form a "small circle". From the perspective of project management, this paper considers the user preference, the knowledge matching degree between the project and experts, the similarity between the projects and the historical performance of experts, and integrates the evaluation expert recommendation algorithm based on the content features, potential topic features, collaborative filtering and user preference, And through the design of experiments, the use of space projects for experimental research, the results show that the proposed algorithm is feasible and practical.

G311;TP391.3

无