Automated content analysis for construction safety: A natural language processing system to extract precursors and outcomes from unstructured injury reports
详细信息 查看全文
- 作者：Antoine J.-P. Tixiera ; antoine.tixier-1@colorado.edu" class="auth_mail" title="E-mail the corresponding author ; Matthew R. Hallowella ; matthew.hallowell@colorado.edu" class="auth_mail" title="E-mail the corresponding author ; Balaji Rajagopalanb ; rajagopalan.balaji@colorado.edu" class="auth_mail" title="E-mail the corresponding author ; Dean Bowmanc ; dean.bowman@bentley.com" class="auth_mail" title="E-mail the corresponding author
- 关键词：Automated content analysis ; Natural language processing ; Text mining ; Knowledge extraction ; Accident ; Injury ; Safety ; Attribute ; Risk ; R
- 刊名：Automation in Construction
- 出版年：2016
- 出版时间：February 2016
- 年：2016
- 卷：62
- 期：Complete
- 页码：45-56
- 全文大小：586 K
文摘
- •
A natural language processing system is proposed.
- •
The system is based on hand-coded rules and dictionaries of keywords.
- •
Injury reports are scanned for 102 attributes and outcomes with over 95% accuracy.
- •
Large databases of unstructured injury reports can now efficiently be leveraged.