PENENTUAN JALUR TERBAIK PADA SMART TRANSPORTATION DALAM SMART CITY MENGGUNAKAN METODE RECURRENT NEURAL NETWORK (RNN) DAN LONG SHORT-TERM MEMORY (LSTM) YANG DIOPTIMASI DENGAN BAYESIAN OPTIMIZATION (RNN-BO DAN LSTM-BO)

SIMATUPANG, DAVID HASIAN and Oklilas, Ahmad Fali (2023) PENENTUAN JALUR TERBAIK PADA SMART TRANSPORTATION DALAM SMART CITY MENGGUNAKAN METODE RECURRENT NEURAL NETWORK (RNN) DAN LONG SHORT-TERM MEMORY (LSTM) YANG DIOPTIMASI DENGAN BAYESIAN OPTIMIZATION (RNN-BO DAN LSTM-BO). Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011381924082.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
	Text RAMA_56201_09011381924082_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (11MB) | Request a copy
	Text RAMA_56201_09011381924082_0015107201_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (707kB)
	Text RAMA_56201_09011381924082_0015107201_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (791kB) | Request a copy
	Text RAMA_56201_09011381924082_0015107201_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_56201_09011381924082_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_09011381924082_0015107201_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (94kB) | Request a copy
	Text RAMA_56201_09011381924082_0015107201_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (98kB) | Request a copy
	Text RAMA_56201_09011381924082_0015107201_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (914kB) | Request a copy

Abstract

The best route determining system is part of the Intelligent Transportation concept where cities that implement this concept are called Smart Cities. The study used an object detection system called You Only Look Once version 8 (YOLOv8) to detect and calculate the number of vehicles based on CCTV recordings. In addition, it uses Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) optimized with Bayesian Optimization (RNN-BO and LSTM-BO) to predict road density conditions. The graph theory algorithm is used to determine the best path. The data set consists of 4224 images of motorcycle vehicles and cars and a road conditions reference table with 5 columns and 320 rows. YOLOv8 produced models with Mean Average Precision (mAP) of 85.4%, with motorcycle detection accuracy of 76% and cars of 86%. And for the accuracy of the readings YOLOv8 gets 78.61%. The RNN model has an accuracy of 50.16% and a reading accuration of 85.75%, while the LSTM model has a accurate of 50.48% and reading accurateness of 86.76%. After optimization using Bayesian Optimization, the accuracy of the RNN-BO model increased to 55.56% and the LSTM-BO to 54.60%. However, there were no changes in the accuracy of the reading, which is 85.75% for RNN-BO and 86.76% for LSTM-BO. In the implementation of Graph Theory for determining the best path, a consistent result is obtained: Route 4 in all road conditions, based on the parameters of road conditions and the mileage that produces the smallest weight.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Penentuan Jalur Terbaik, Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Bayesian Optimization, Graph Theory
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning T Technology > T Technology (General) > T1-995 Technology (General)
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	David Hasian Simatupang
Date Deposited:	26 Jul 2023 02:14
Last Modified:	26 Jul 2023 02:14
URI:	http://repository.unsri.ac.id/id/eprint/121444

Actions (login required)

View Item