https://hdl.handle.net/20.500.12809/182 2026-04-04T07:16:28Z 2026-04-04T07:16:28Z Gülbaş, Gülşah Çetin, Gürcan https://hdl.handle.net/20.500.12809/10983 2023-09-27T10:35:13Z 2023-01-01T00:00:00Z

Lifetime Optimization of the LEACH Protocol in WSNs with Simulated Annealing Algorithm Gülbaş, Gülşah; Çetin, Gürcan The lifetime of a Wireless Sensor Network (WSN) is determined by its energy restriction. One of the conventional techniques used to maintain network connectivity is the utilization of the LEACH routing protocol. LEACH is based on clustering, and the process of choosing a Cluster Head (CH) in each round is based on chance. Consequently, it remains unclear whether the best CH is selected for each round. In this study, two approaches based on the Simulated Annealing (SA) algorithm are described to minimize energy losses of the nodes and improve the lifetime of the WSN utilizing the LEACH routing protocol. In both techniques, the residual energies at the nodes, as well as their distances from each other, are taken into consideration when determining the CHs. The efficiency of the presented approaches has been evaluated for networks with 10, 25, 50 and 100 sensors in terms of consumed energy, total data packets received by the Base Station (BS), the number of active/dead nodes, and the average energy per sensor. According to the findings, the PSCH-SA technique yields the most favorable results in networks with 10 sensors, while the LEACH-SA protocol demonstrates superior performance in WSNs with 25 or more sensors.

2023-01-01T00:00:00Z Koçak, Hilal Çetin, Gürcan https://hdl.handle.net/20.500.12809/10820 2023-08-01T06:53:35Z 2023-01-01T00:00:00Z

The Diagnosis of Diabetes Mellitus with Boosting Methods Koçak, Hilal; Çetin, Gürcan In addition to the damage, it can cause to various organs, diabetes mellitus (DM) also increases a person's risk of developing other serious health conditions. These can include heart disease, stroke, and nerve damage. Furthermore, DM is a leading cause of blindness and kidney failure. However, with proper management and treatment, many of the complications of DM can be prevented or delayed. Thus, early detection and treatment of DM are crucial. With the advancement of machine learning technology, new opportunities have emerged in the field of medicine. Many disease detection research relies on machine learning techniques, with a particular emphasis on boosting algorithms. Boosting algorithms are used to improve the accuracy of predictions made by other weak models such as decision trees. Using knowledge discovery methods, boosting algorithms are examined and compared on a diabetes dataset in this study. The performance of the boosting algorithms is evaluated by generating ROC curves and comparing average accuracy values. When the study's results were evaluated in terms of precision, Gradient Boosting, AdaBoost, CatBoost, LightGBM, and XGBoost algorithms gives success rates of %85, %83, %88, %86, and %87, respectively.

2023-01-01T00:00:00Z Siddiqi, Muhammad Hameed Alruwaili, Madallah Tarımer, İlhan Karadağ, Buse Cennet Alhwaiti, Yousef https://hdl.handle.net/20.500.12809/10763 2024-12-05T13:54:44Z 2023-01-01T00:00:00Z

Development of a Smart Signalization for Emergency Vehicles Siddiqi, Muhammad Hameed; Alruwaili, Madallah; Tarımer, İlhan; Karadağ, Buse Cennet; Alhwaiti, Yousef As the population increases, the number of motorized vehicles on the roads also increases. As the number of vehicles increases, traffic congestion occurs. Traffic lights are used at road junctions, intersections, pedestrian crossings, and other places where traffic needs to be controlled to avoid traffic chaos. Due to traffic lights installed in the city, queues of vehicles are formed on the streets for most of the day, and many problems arise because of this. One of the most important problems is that emergency vehicles, such as ambulances, fire engines, police cars, etc., cannot arrive on time despite traffic priorities. Emergency vehicles such as hospitals and police departments need to reach the scene in a very short time. Time loss is a problem that needs to be addressed, especially for emergency vehicles traveling in traffic. In this study, ambulances, fire brigades, police, etc., respond to emergencies. A solution and a related application have been developed so privileged vehicles can reach their target destination as soon as possible. In this study, a route is determined between the current location of an emergency vehicle and its target location in an emergency. Communication between traffic lights is provided with a mobile application developed specifically for the vehicle driver. In this process, the person controlling the lights can turn on the traffic lights during the passage of vehicles. After the vehicles with priority to pass passed, traffic signaling was normalized via the mobile application. This process was repeated until the vehicle reached its destination.

2023-01-01T00:00:00Z Sağbaş, Ensar Arif Korukoğlu, Serdar Ballı, Serkan https://hdl.handle.net/20.500.12809/10720 2023-05-26T08:00:31Z 2023-01-01T00:00:00Z

Real-time stress detection from smartphone sensor data using genetic algorithm-based feature subset optimization and k-nearest neighbor algorithm Sağbaş, Ensar Arif; Korukoğlu, Serdar; Ballı, Serkan Stress is the mood of pressure and tension that a person feels. Usually, when the pressure on an individual decrease, the body begins to stabilize the state and calm down. Hence, stress detection in real-time is a critical duty in medical systems. However, acquiring physiological data requires additional equipment and is difficult for users to carry with them at all times. Depending on this problem, it is possible to detect stress through behavioral data. Smartphones are devices that provide various behavioral data that people use constantly throughout the day. In this study, a real-time stress detection system based on soft keyboard typing behaviors was developed with the data obtained from linear acceleration, gravity, gyroscope sensors, and a touchscreen panel of the smartphone. 172 attributes were extracted from the raw sensor data. However, such a high number of dimensions could negatively affect the performance of machine learning algorithms. To address this problem, the number of features was reduced by various techniques such as filter-based methods and standard binary-code chromosome Genetic Algorithm as a contribution to this study. Then, writing behaviors were classified with the commonly used machine learning methods namely, C4.5, kNN, and Bayesian Networks. As a result of the experiments, the best classification was obtained from the kNN method using the features selected by the Genetic Algorithm with a classification accuracy of 89.61% and F-Measure of 0.9052. Another contribution of this study is that a mobile service and a relaxation application were developed for stress detection and to reduce stress levels using the selected feature vector.

2023-01-01T00:00:00Z