Combining GPS and sensors to determine mode of transportation
Chandrasegaran, Gajalakshan (2023)
Chandrasegaran, Gajalakshan
2023
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2023052212712
https://urn.fi/URN:NBN:fi:amk-2023052212712
Tiivistelmä
The purpose of this thesis project was to identify different modes of transportation by making use of GPS data obtained from mobile devices, in contrast to other research of a similar nature.
GPS-based methods use the location data provided by GPS sensors, whereas sensor-based methods use data from other sensors such as accelerometers, gyroscopes, and magnetometers. By combining both strategies, the strengths of each can be leveraged to produce more precise results. These combined methods are evaluated using a variety of metrics, such as precision, recall and F1 score.
The findings of this thesis project indicate that the machine learning model devised in this research is proficient in precisely categorizing various transportation modes by utilizing GPS and sensor data. Therefore, it can be inferred that the model is effective. The model exhibited a notable degree of precision, as evidenced by an average accuracy rating of 88%, signifying its ability to accurately discern the mode of transportation in the majority of instances.
GPS-based methods use the location data provided by GPS sensors, whereas sensor-based methods use data from other sensors such as accelerometers, gyroscopes, and magnetometers. By combining both strategies, the strengths of each can be leveraged to produce more precise results. These combined methods are evaluated using a variety of metrics, such as precision, recall and F1 score.
The findings of this thesis project indicate that the machine learning model devised in this research is proficient in precisely categorizing various transportation modes by utilizing GPS and sensor data. Therefore, it can be inferred that the model is effective. The model exhibited a notable degree of precision, as evidenced by an average accuracy rating of 88%, signifying its ability to accurately discern the mode of transportation in the majority of instances.