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

作　　者：杨茜 朱佳悦 常远[2] YANG Qian;ZHU Jiayue;CHANG Yuan(China Architecture Design&Research Group,Beijing 100044,China;School of Management Science and Engineering,Central University of Finance and Economics,Beijing 100081,China)

机构地区：[1]中国建筑设计研究院有限公司,北京100044 [2]中央财经大学管理科学与工程学院,北京100081

出　　处：《建筑经济》2024年第S01期241-245,共5页Construction Economy

基　　金：国家自然科学基金项目“代谢视角下城市建筑脱碳能力形成机理、演化规律及提升路径研究”(72071220)。

摘　　要：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.

关 键 词：成本估算 建筑特征 深度学习 信息抽取 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TU238.2[自动化与计算机技术—控制科学与工程]