Publication: Balancing performance and computational cost for online petition processing
Balancing performance and computational cost for online petition processing
Date
Date
Date
Citations
Siswanto, C. K., Kovacs, M., & Serdült, U. (2025). Balancing performance and computational cost for online petition processing. 86–92. https://doi.org/10.1145/3728985.3729000
Abstract
Abstract
Abstract
This research explores cost-effective approaches to classifying online petitions by comparing two deep learning models: BERT base uncased and DistilBERT. The data used is from a UK government online petition website. The impact of different fine-tuning strategies, such as training all layers versus freezing certain layers in the machine learning architecture, is investigated to analyze their effects on classification performance and efficiency. Experiments show that DistilBERT is only 2% less accurate than BERT, but significantly redu
Metrics
Downloads
Views
Citations
Additional indexing
Creators (Authors)
- Siswanto, Christopher Kevin
- Kovacs, Mate
Event Title
Event Title
Event Title
Event Location
Event Location
Event Location
Event Start Date
Event Start Date
Event Start Date
Event End Date
Event End Date
Event End Date
Publisher
Publisher
Publisher
Page range/Item number
Page range/Item number
Page range/Item number
Page end
Page end
Page end
Item Type
Item Type
Item Type
Language
Language
Language
Date available
Date available
Date available
ISBN or e-ISBN
ISBN or e-ISBN
ISBN or e-ISBN
OA Status
OA Status
OA Status
Free Access at
Free Access at
Free Access at
Publisher DOI
Metrics
Downloads
Views
Citations
Citations
Siswanto, C. K., Kovacs, M., & Serdült, U. (2025). Balancing performance and computational cost for online petition processing. 86–92. https://doi.org/10.1145/3728985.3729000
