TY - JOUR
T1 - Customized AutoML
T2 - An Automated Machine Learning System for Predicting Severity of Construction Accidents
AU - Toğan, Vedat
AU - Mostofi, Fatemeh
AU - Ayözen, Yunus Emre
AU - Behzat Tokdemir, Onur
N1 - Publisher Copyright:
© 2022 by the authors.
PY - 2022/11
Y1 - 2022/11
N2 - Construction companies are under pressure to enhance their site safety condition, being constantly challenged by rapid technological advancements, growing public concern, and fierce competition. To enhance construction site safety, literature investigated Machine Learning (ML) approaches as risk assessment (RA) tools. However, their deployment requires knowledge for selecting, training, testing, and employing the most appropriate ML predictor. While different ML approaches are recommended by literature, their practicality at construction sites is constrained by the availability, knowledge, and experience of data scientists familiar with the construction sector. This study develops an automated ML system that automatically trains and evaluates different ML to select the most accurate ML-based construction accident severity predictors for the use of construction professionals with limited data science knowledge. A real-life accident dataset is evaluated through automated ML approaches: Auto-Sklearn, AutoKeras, and customized AutoML. The investigated AutoML approaches offer higher scalability, accuracy, and result-oriented severity insight due to their simple input requirements and automated procedures.
AB - Construction companies are under pressure to enhance their site safety condition, being constantly challenged by rapid technological advancements, growing public concern, and fierce competition. To enhance construction site safety, literature investigated Machine Learning (ML) approaches as risk assessment (RA) tools. However, their deployment requires knowledge for selecting, training, testing, and employing the most appropriate ML predictor. While different ML approaches are recommended by literature, their practicality at construction sites is constrained by the availability, knowledge, and experience of data scientists familiar with the construction sector. This study develops an automated ML system that automatically trains and evaluates different ML to select the most accurate ML-based construction accident severity predictors for the use of construction professionals with limited data science knowledge. A real-life accident dataset is evaluated through automated ML approaches: Auto-Sklearn, AutoKeras, and customized AutoML. The investigated AutoML approaches offer higher scalability, accuracy, and result-oriented severity insight due to their simple input requirements and automated procedures.
KW - automated machine learning (AutoML)
KW - construction accident
KW - multi-level severity prediction
KW - risk assessment
UR - http://www.scopus.com/inward/record.url?scp=85143539410&partnerID=8YFLogxK
U2 - 10.3390/buildings12111933
DO - 10.3390/buildings12111933
M3 - Article
AN - SCOPUS:85143539410
SN - 2075-5309
VL - 12
JO - Buildings
JF - Buildings
IS - 11
M1 - 1933
ER -