The Use of Machine Learning Methods to Determine Risk Zones for Construction Disasters Caused by Wind - A Case Study in Poland
Year: 2024
Type: article
Abstract: The increasing availability and use of artificial intelligence (AI) methods and algorithms have led to their widespread use i n analyses aimed at identifying the decisive factors that influence the course of a studied phenomenon or process. AI algorithm s include a wide range of methods. They can be used together or separately. The article describes the use of two Machine Learni ng (ML) methods, PCA and k - means, to identify parameters that may increase or decrease the risk of construction disasters caused by strong winds in Poland. The analysis was conducted using a unique dataset on construction disasters in Europe, sourced from t he General Office of Construction Supervision in Poland for the period 1995 - 2019. The occurrence of disasters was categorised by voivodeship and cause, with information provided on the number of people injured. ML analyses were conducted to determine whether land development, population density, and weather factors, such as wind, have an impact on the number of recorded disasters .
Source: Inżynieria Mineralna
Author: Ireneusz Laks
Institution: Poznań University of Life Sciences
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4413627950