Machine Learning-Aided Prediction of Pile Behaviour: The Role of Data Quality
- Publisher:
- Springer Nature
- Publication Type:
- Chapter
- Citation:
- Proceedings of the 5th International Conference on Geotechnics for Sustainable Infrastructure Development, 2024, 395, pp. 515-526
- Issue Date:
- 2024-01-01
Closed Access
Filename | Description | Size | |||
---|---|---|---|---|---|
978-981-99-9722-0_35 (1).pdf | Published version | 705.5 kB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
Machine learning (ML), a data-based approach, has recently emerged as an effective method to predict the behaviour of pile foundation. However, the assessment of data quality is usually carried out with a lack of thorough consideration and/or robust methodology. This paper emphasizes key aspects to evaluate data quality with reference to the most common ML algorithms and practice of pile foundation. An investigation into the data randomness and uniformity during selection of data sets for training and testing ML model based on the Artificial Neural Network (ANN) is carried out to demonstrate the possible influence of inadequate data quality on the predicted outcome. The results imply the utmost importance of having high-quality data for building cost-effective and reliable ML model.
Please use this identifier to cite or link to this item: