IMPLEMENTASI ALGORITMA K-NEAREST NEIGHBOR DAN BAYESIAN OPTIMIZATION DALAM MENENTUKAN JALUR TERBAIK SMART TRANSPORTATION PADA SMART CITY

SEPRIYANI, YEPI and Oklilas, Ahmad Fali (2023) IMPLEMENTASI ALGORITMA K-NEAREST NEIGHBOR DAN BAYESIAN OPTIMIZATION DALAM MENENTUKAN JALUR TERBAIK SMART TRANSPORTATION PADA SMART CITY. Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011181924022.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_56201_09011181924022_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_56201_09011181924022_0015107201_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (774kB)
	Text RAMA_56201_09011181924022_0015107201_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (293kB) | Request a copy
	Text RAMA_56201_09011181924022_0015107201_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (376kB) | Request a copy
	Text RAMA_56201_09011181924022_0015107201_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (126kB) | Request a copy
	Text RAMA_56201_09011181924022_0015107201_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (171kB) | Request a copy
	Text RAMA_56201_09011181924022_0015107201_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011181924022_0015107201_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

The increase in the number of vehicles and the urban population has led to high levels of traffic congestion in urban areas. The focus of this research is to find the best alternative routes using various methods and algorithms to detect objects and determine the optimal paths in the city of Palembang. Object detection is carried out using the YOLOv3 method through video recording results, with an overall mean average precision (mAP) of 75.63%. Specifically, the mAP is 71.28% for the motorbike class and 79.99% for the car class. The K-Nearest Neighbor method is used to determine road density conditions, whether it is smooth, moderate, or congested, based on the number of motorbikes and cars, lane count, and road length. Subsequently, an optimization process is conducted using Bayesian optimization, resulting in a model accuracy increase from 85.93% to 93.75%. The outcomes, which include road condition classifications at each intersection, are then processed by the Breadth First Search algorithm. The decision is made that route 5 is the dominant choice for the best path. Out of the 12 tested time instances, 10 indicated that route 5 was the best option. The determination of this route is based on the magnitude of the produced weight. Route 5 tends to have a smaller weight factor due to its prevalent "smooth" road conditions and also represents the shortest distance compared to other routes.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	YOLOv3, K-Nearest Neighbor, Bayesian Optimization, Breadth First Search, Jalur Terbaik
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Yepi Sepriyani
Date Deposited:	26 Sep 2023 06:21
Last Modified:	26 Sep 2023 06:21
URI:	http://repository.unsri.ac.id/id/eprint/129178

Actions (login required)

View Item