摘要
[目的/意义]用户模型是非物质文化遗产(以下简称非遗)资源自适应推送系统的核心环节,其描述了用户诸多的个性化特征和不同的需求与目标,是系统进行领域本体资源调用和适当推送策略选择的依据。[方法/过程]首先,从用户的基本信息、访问风格和知识状态3个方面设计用户模型;其次,借助所罗门风格量表初始化用户访问风格模型,同时确定和采集用户有效的访问行为数据,并利用贝叶斯网络算法挖掘用户行为数据,从而推测出用户的真实访问风格;最后,明确不同用户的需求与目标,采用知识多重覆盖模型初始化用户知识状态模型,并通过单个知识点测试评估用户的知识掌握状态和能力水平的变化,以动态调整非遗知识内容推送。[结果/结论]经实践,该用户模型有助于不同用户对非遗知识的学习、交流和掌握,能解决非遗资源个性化推送中出现的诸多问题。
[ Purpose/Significance] User model is the core of the adaptive knowledge pushing system of Intangible Cultural Heritage re-sources. It describes user's personalized features and different knowledge needs and goals, and it's the basis of the system to invoke domainontology resources and choose suitable pushing strategies. [Method/Process]Firstly, this paper designs a user model from user's basic in-formation , accessing style and state of knowledge. Secondly, it initializes user 爷 s accessing style model with Solomon style scale, confirmsand collects user's valid accessing behavior data, and digs these data with Bayesian Network algorithm, to deduce the real accessing style.Finally, it defines different user's demands and goals, initializes user's knowledge state model with multiple coverage model of knowledge,and tests and evaluates the changing of mastery state and ability level with single knowledge point, to dynamically adjust the pushing of in-tangible cultural heritage knowledge. [Result/Conclusion] As practice confirms that this user model can benefit users in their learning,communicating and mastering process of Intangible Cultural Heritage, and it can help to solve many problems in personalized pushing.
作者
庄文杰
谈国新
侯西龙
李莎
Zhuang Wenjie;Tan Guoxin;Hou Xilong;Li Sha(School of Educational Information Technology,Central China Normal University,Wuhan 430079;National Research Center o f Cultural Industries,Central China Normal University,Wuhan 430079)
出处
《情报杂志》
CSSCI
北大核心
2017年第11期106-113,105,共9页
Journal of Intelligence
基金
国家社会科学基金重点项目"科技提高我国文化软实力和竞争力的路径研究"(编号:13A2D008)研究成果之一
关键词
非物质文化遗产
自适应推送
用户模型
访问风格
知识状态
intangible cultural heritage adaptive knowledge pushing user model accessing style knowledge state