نوع مقاله : علمی - پژوهشی
نویسندگان
1 دانشکده مهندسی نقشه برداری و اطلاعات مکانی، دانشکدگان فنی، دانشگاه تهران
2 دانشکده مهندسی معدن، دانشگاه صنعتی سهند
3 دانشکده مهندسی نقشه برداری و اطلاعات مکانی، دانشکگان فنی، دانشگاه تهران
چکیده
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Highways constitute one of the vital linear elements in a country's transportation system, and their significance has grown steadily with the expansion of urban life. Consequently, access to up-to-date and efficient information regarding these linear elements plays a crucial role in managing the transportation systems of nations. Volunteered Geographic Information is generated by users who input their experiential or local knowledge of a specific location into a spatial database. Statistical approaches aid in understanding the trends of participatory behaviors. This study examines the patterns of contributors' involvement when mapping highways and roads in the OpenStreetMap database in Tehran. The proposed approach evaluates how the characteristics of a road impact its classification. Highway classes are classified using a random forest classifier, and the inferred mapping trends are analyzed based on the classification. Linear data from Tehran city are utilized for the upcoming experiment, and the classification algorithm is applied with the extraction parameters that yielded the highest classification accuracy. Results demonstrate that the best combination of extraction parameters, including the azimuth of highways, geodesic length of lines, distance from the first mapped point to the nearest street, and the kernel density of the first points, achieves an F-Score of 71%. An examination of the semantic parameters such as the highway name and user name indicates their ineffectiveness on classification accuracy. After assessing the importance of each selected parameter, the azimuth parameter is identified as the most influential factor in the classification of highway types in the behavior of users participation.
کلیدواژه‌ها [English]
)