Abstract
Driver behaviour modelling is a critical field that addresses complex and dynamic driving behaviours on roads with the goal of enhancing road safety, reducing air pollution, and improving vehicle performance. Recent advancements in sensor technology and machine learning (ML) techniques have facilitated the capture and analysis of driver behaviour patterns. Nonetheless, the efficacy of ML models heavily relies on the quality of the data used. Therefore, developing feature extraction techniques that provide high-quality inputs is crucial. In this paper, we conceptualised, implemented, and evaluated a novel feature model called Changepoint-based Statistical Feature (C-bSF). Initially, we extracted various statistical functions from raw sensor data, which were then aggregated using lagging windows. Following this, a changepoint detection method was used to derive the C-bSF feature. We compared the performance metrics of this new approach with other feature extraction methods, demonstrating the superiority of C-bSF in driver behaviour classification tasks across three datasets.
Original language | English |
---|---|
Title of host publication | PRICAI 2024: Trends in Artificial Intelligence |
Subtitle of host publication | 21st Pacific Rim International Conference on Artificial Intelligence, PRICAI 2024 Kyoto, Japan, November 18–24, 2024 |
Editors | Rafik Hadfi, Alok Sharma, Quan Bai, Patricia Anthony, Takayuki Ito |
Publisher | Springer |
Pages | 231-242 |
Number of pages | 12 |
Volume | 15281 |
ISBN (Electronic) | 9789819601165 |
ISBN (Print) | 9789819601158 |
DOIs | |
Publication status | Published - 2025 |
Event | 21st Pacific Rim International Conference on Artificial Intelligence 2024: PRICAI2024 - Kyoto University, Kyoto, Japan Duration: 18 Nov 2024 → 24 Nov 2024 https://www.pricai.org/2024/ https://www.dropbox.com/scl/fi/lreaqzj40nt7piqcfcuhr/PRICAI-2024-Whole-Program-16112024-Ver-5.pdf?rlkey=ga7dbh7us284ld3tybao8n66x&e=2&st=ru4ds4m6&dl=0 (Program) https://www.pricai.org/2024/index.php/programs/accepted-papers (Proceedings) https://link.springer.com/content/pdf/bfm:978-981-96-0116-5/1 (Front matter) |
Publication series
Name | Lecture Notes in Artificial Intelligence |
---|---|
Publisher | Springer |
Volume | 15281 |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | 21st Pacific Rim International Conference on Artificial Intelligence 2024 |
---|---|
Country/Territory | Japan |
City | Kyoto |
Period | 18/11/24 → 24/11/24 |
Other | The Pacific Rim International Conference on Artificial Intelligence (PRICAI) is an annual international event which concentrates on AI theories, technologies and their applications in the areas of scientific, social, and economic importance for countries in the Pacific Rim. In the past, the conferences have been held in Nagoya (1990), Seoul (1992), Beijing (1994), Cairns (1996), Singapore (1998), Melbourne (2000), Tokyo (2002), Auckland (2004), Guilin (2006), Hanoi (2008), Daegu (2010), Kuching (2012), Gold Coast (2014), Phuket (2016), Nanjing (2018), Fiji (2019), Yokohama (2020, online), Hanoi (2021, online), Shanghai (2022, hybrid) and Jakarta (2023, hybrid). PRICAI 2024 will be held in person in Kyoto, Japan. The Program Committee invites technical papers on substantial, original, and unpublished research in all aspects of Artificial Intelligence. PRICAI-2024 aims to bring together researchers, practitioners, educators and users in AI and related communities for in-depth intellectual exchanges, research cooperation and professional development. |
Internet address |
|