摘要
目前工程建设中产生的大量信息仍主要以非结构化文档的形式进行存储,本文研究非结构化信息与高度结构化的BIM模型数据的关联机制,以实现基于BIM的非结构化信息的查询与提取,提高其使用效率。本文提出基于文档标题和文档内容相结合的关联机制,搜索与特定文档相关的模型实体,并根据计算的相关度进行排序,经用户确认后建立文档与模型实体之间的关联。实例验证表明,如果模型实体命名规范,文档表述与之较为匹配,本方法可以召回所有相关实体,且相关实体均出现在列表的前方,关联后的文档可以基于BIM实体动态查询,提高了非结构化信息的利用效率。
Large amount of information in construction projects is still stored in unstructured documents nowadays. This research focuses on the correlation mechanism between unstructured information with highly structured BIM model,realizing BIM-based retrieval of unstructured information and improving the using efficiency. Titlebased and content-based automatic correlation mechanisms are provided,capable of searching correlated entities through selected document,calculating correlation index and ranking and establishing the correlation after users' confirmation. A case-study shows that if entities are well named and expressions in the document match the model,this method is able to recall all the correlated entities,ranking at the top of the list. Correlated document can also be queried dynamically based on BIM model.
出处
《土木建筑工程信息技术》
2015年第3期16-21,共6页
Journal of Information Technology in Civil Engineering and Architecture
基金
国家自然科学基金面上项目(51278274)
清华大学(土水学院)-广联达BIM联合研究中心基础研究课题专项
关键词
BIM
非结构化信息
工程文档
关联机制
BIM
Unstructured Information
Project Documents
Correlation Relationship