000			02801nam a22002417a 4500
003			FT8786
005			20251107150451.0
008			251107b ||||| |||| 00| 0 eng d
041			_aengtag
050			_aT58.5 B37 2025
082			_a.
100	1		_a Barredo, Armela Rose D.; Conde, Mhikaela R.; Lasay, Francheska Anne M.
245			_aThe development of a speed anomaly detection utilizing isolation forest algorithm as evidence for unforeseen road incidents
264		1	_a. _b. _cc2025
300			_bUndergraduate Thesis: (Bachelor of Science in Information Technology) - Pamantasan ng Lungsod ng Maynila, 2025
336			_2text _atext _btext
337			_2unmediated _aunmediated _bunmediated
338			_2volume _avolume _bvolume
505			_aABSTRACT: Overspeeding is a critical cause of road accidents, particularly for motorcycle riders, as it compromises vehicle control, increases crash severity, and reduces reaction time. Many riders struggle to monitor their speed effectively, relying on manual speedometer checks that can be distracting and lead to potential accidents. Unlike high-end vehicles with built-in alarm systems, most motorcycles lack an intuitive and accessible warning mechanism to alert riders when they exceed the speed limit. Additionally, the absence of an automated system for logging road incidents limits the ability to track critical data such as location, time, and driving patterns, making accident analysis and prevention more challenging. Furthermore, riders often remain unaware of irregularities in their driving patterns, as there is no existing mechanism to detect sudden speed fluctuations or erratic acceleration that could indicate unsafe riding behavior. The urgency of this issue cannot be overstated. With motorcycle being a prevalent mode of transportation, especially in urban and developing areas, the risks associated with overspeding pose a severe threat to public safety. To address these issues, this project develops an Arduino-based motorcycle speed detection system integrated with a mobile application for real-time data logging and anomaly detection. The system utilizes the Isolation Forest algorithm to identify speed-related anomalies and enhance road safety by alerting rides through a speed alarm when speed limits are breached while systematically recording road incidents. Experimental results showed that the system demonstrated strong precision in detecting anomalies with minimal false alarms, though some cases were still missed. The overall accuracy suggests reliability in distinguishing normal and anomalous driving events, but further refinement and testing under diverse conditions are recommended to enhance the effectiveness and adaptability.
526			_aF
655			_aacademic writing
942			_2lcc _cMS
999			_c37048 _d37048