基于深度学习的成本估算决定性建筑特征自动抽取方法研究

Research on Automatic Extraction of Deterministic Building Features for Cost Estimation Based on Deep Learnins

在建筑业数智化转型背景下,自动识别建筑成本估算的决定性建筑特征,助力实现高精度建筑成本管理,具有重要理论和实践意义。本文解析建筑工程项目初步设计说明复杂文本特征,应用深度学习算法揭示建筑特征在文本中的分布模式,提出决定成本估算的建筑特征自动抽取方法,并基于海量文本验证抽取方法的有效性。

Within the digital transformation context of construction industry,automatic identification of deterministic building features for cost estimation plays a significant role in achieving high-precision cost management.This paper analyzes the complex textual features of the Preliminary Design Specification documents of construction projects,applies deep learning algorithms to reveal the distribution patterns of building features within documents,and proposes an automatic method for extraction of building features that determine cost estimation.Finally,the effectiveness of the proposed extraction method was experimentally validated based on a large amount of textual data.