Deep learning-based chatbot by natural language processing for supportive risk management in river dredging projects

作     者：Chou, Jui-Sheng Chong, Pei-Lun Liu, Chi-Yun 

作者机构：Department of Civil and Construction Engineering National Taiwan University of Science and Technology Taipei Taiwan UTRUST Technology Co. Ltd Taipei Taiwan 

出 版 物：《Engineering Applications of Artificial Intelligence》 (Eng Appl Artif Intell)

年 卷 期：2024年第131卷

核心收录：

学科分类：0710[理学-生物学] 1202[管理学-工商管理] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 081501[工学-水文学及水资源] 0837[工学-安全科学与工程] 0815[工学-水利工程] 0903[农学-农业资源与环境] 0835[工学-软件工程] 0836[工学-生物工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：The authors would like to thank the National Science and Technology Council  Taiwan   for financially supporting this research under contracts NSTC 108-2221-E-011-003-MY3 and 110-2221-E-011-080-MY3 

主　　题：Expert systems 

摘      要：Dredging projects play a vital role in water conservation, especially for flood control, but they often encounter natural and human-induced risks. This study addresses these challenges by creating a comprehensive knowledge base for dredging risks. This knowledge base serves several essential functions: it aids in immediate and thorough risk prevention, provides practical risk management recommendations, and assists dredging personnel in making informed decisions when dealing with risks. Furthermore, the study introduces an advanced deep-learning model that enhances access to critical information. Users can quickly find relevant risk prevention strategies by inputting keywords. The model s distinctive feature is its ability to analyze unforeseen risks through natural language processing, predict the potential impact and frequency of unexpected events, and propose appropriate risk response measures. This empowers dredging personnel to conduct preliminary risk assessments confidently. The deep learning model seamlessly integrates into LINE s communication platform, creating a river dredging project risk knowledge chatbot. This user-friendly chatbot is accessible to personnel at all project stages, enabling real-time interaction. It offers a practical way to develop and implement risk prevention plans and response measures. In summary, this research presents an innovative approach that enhances the efficiency and safety of dredging projects. By merging a knowledge base with cutting-edge technology and real-time communication, it equips dredging personnel to manage known and unforeseen risks proficiently, ultimately contributing to the success of water conservancy and flood control projects. © 2023 Elsevier Ltd