工程资料的自动分类和智能搜索技术探析

Analysis on Automated Classification and Intelligent Search Method of Engineering Documents

由于缺乏引入针对性的文本分析方法,企业级工程资料的快速积累,带来了搜索获取所需信息的困难。为解决该问题,首先将多种格式的工程资料提取关键词,转化为文档向量,表达主要语义内容;然后采用聚类找到主要的文档类型,并基于此训练了一种施工文档自动分类的神经网络模型,以便将工程资料组织起来进行分类检索。此外,还基于语义分析,提出一种智能化的搜索和相似推荐的算法,实现了快速获取所需的工程技术资料的效果。

Due to the lack of targeted text analysis methods, the rapid accumulation of enterprise engineering data brings difficulties in searching for the required information.To solve this problem, keywords are extracted and transformed from engineering files of various formats into file vectors to express the main semantic content.Then, the main categories of documents are found by clustering.A neural network model for automatic classification of construction documents is trained to organize engineering data for classification and retrieval.In addition, Based on semantic analysis, an intelligent search and similarity recommendation algorithm is proposed, which can support efficient data requirement and research.