Yazar "Dokuz, Ahmet Sakir" seçeneğine göre listele
Listeleniyor 1 - 17 / 17
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Anomaly Detection in Region Mobility Utilization Using Daily Taxi Trajectory Dataset(Springer International Publishing Ag, 2022) Dokuz, Yesim; Dokuz, Ahmet SakirAnomaly detection in urban big datasets is getting wide attention with the presence of different and various urban big data sources. Urban anomaly detection is an important application area because discovered anomalies in urban areas would provide essential information about urban areas and their utilization, especially human mobility analytics and traffic condition monitoring. In the literature, there are several studies performed for urban anomaly detection using taxi trajectory datasets, such as events detection, regional urban anomaly detection and traffic incident detection. In this study, anomaly detection in regional mobility utilization of daily taxi trajectory datasets is performed based on DBSCAN clustering algorithm. A new algorithm and a threshold value are proposed to detect taxi regions as normal and anomalous for both incoming and outgoing taxi trip records. Experiments are performed on New York taxi trajectory big dataset and the experimental results show that proposed algorithm is efficient on detecting regions as normal or anomalous based on daily taxi trip record counts.Öğe Cloud Computing-Based Socially Important Locations Discovery on Social Media Big Datasets(World Scientific Publ Co Pte Ltd, 2020) Dokuz, Ahmet Sakir; Celik, MeteSocially important locations are places which are frequently visited by social media users in their social media lifetime. Discovering socially important locations provides valuable information, such as which locations are frequently visited by a social media user, which locations are common for a social media user group, and which locations are socially important for a group of urban area residents. However, discovering socially important locations is challenging due to huge volume, velocity, and variety of social media datasets, inefficiency of current interest measures and algorithms on social media big datasets, and the need of massive spatial and temporal calculations for spatial social media analyses. In contrast, cloud computing provides infrastructure and platforms to scale compute-intensive jobs. In the literature, limited number of studies related to socially important locations discovery takes into account cloud computing systetns to scale increasing dataset size and to handle massive calculations. This study proposes a cloud-based socially important locations discovery algorithm of Cloud SS-ILM to handle volume and variety of social media big datasets. In particular, in this study, we used Apache Hadoop framework and Hadoop MapReduce programming model to scale dataset size and handle massive spatial and temporal calculations. The performance evaluation of the proposed algorithm is conducted on a cloud computing environment using Turkey Twitter social media big dataset. The experimental results show that using cloud computing systems for socially important locations discovery provide much faster discovery of results than classical algorithms. Moreover, the results show that it is necessary to use cloud computing systems for analyzing social media big datasets that could not be handled with traditional stand-alone computer systems. The proposed Cloud SS-ILM algorithm could be applied on many application areas, such as targeted advertisement of businesses, social media utilization of cities for city planners and local governments, and handling emergency situations.Öğe Data-driven energy consumption prediction of a university office building using machine learning algorithms(Pergamon-Elsevier Science Ltd, 2024) Yesilyurt, Hasan; Dokuz, Yesim; Dokuz, Ahmet SakirRedundant consumption of energy in buildings is an important issue that causes increasing problems of climate change and global warming in the world. Therefore, it is necessary to develop efficient energy management approaches in buildings. Accurate prediction of energy consumption plays an important role to obtain energyefficient buildings. Data-driven methods gained attention for estimation of energy consumption in buildings which would provide more accurate prediction results. In this study, hourly energy consumption prediction is performed on a university office building to increase energy efficiency in the building using machine learning algorithms. A new parameter is proposed, air conditioning demand, to improve accuracy of the algorithms. Moreover, temporal parameters, i.e. day of week, month of year, and hour of day, were used along with meteorological parameters to improve prediction performance of the algorithms. Experimental results show that hourly energy consumption of the building could be predicted using machine learning algorithms with high performance. When the results were analysed, Deep Neural Network (DNN) achieved better performance among other alternative algorithms. The average values of R2, RMSE and MAPE for DNN were 0.959, 4.796 kWh, and 5.738 %, respectively. Also, the addition of proposed air conditioning demand parameter provided improved performance to the algorithms.Öğe Dense Region Detection of Streaming Social Media Datasets using Blockchain-based Secure Computation(Institute of Electrical and Electronics Engineers Inc., 2022) Ecemis, Alper; Dokuz, Ahmet Sakir; Celik, MeteThe popularity of blockchain technology is increasing day by day, and it has a widespread usage as a reliable database structure. Blockchain has been utilized across a wide range of industries, such as IoT, social media, robotics, government, supply chain, and healthcare. As the usage of blockchain is emerged in social media studies, the problem of storing and analyzing streaming social media data in a safe, anonymous, and reliable manner is emerged. Within the scope of this problem, in this study, the adaptation of the streaming social media dataset to the blockchain systems and analysis of this data are carried out. A new algorithm of Dense Region Detection Algorithm (DRDA) is proposed to detect dense social regions of urban areas. Streaming data are collected from New York City, and the proposed DRDA is performed by miners to determine the dense regions of New York. As a result, dense regions with intensive tweet sharing in New York are detected and discussed. © 2022 IEEE.Öğe Discovering socially important locations of social media users(Pergamon-Elsevier Science Ltd, 2017) Dokuz, Ahmet Sakir; Celik, MeteSocially important locations are places that are frequently visited by social media users in their social media life. Discovering socially interesting, popular or important locations from a location based social network has recently become important for recommender systems, targeted advertisement applications, and urban planning, etc. However, discovering socially important locations from a social network is challenging due to the data size and variety, spatial and temporal dimensions of the datasets, the need for developing computationally efficient approaches, and the difficulty of modeling human behavior. In the literature, several studies are conducted for discovering socially important locations. However, majority of these studies focused on discovering locations without considering historical data of social media users. They focused on analysis of data of social groups without considering each user's preferences in these groups. In this study, we proposed a method and interest measures to discover socially important locations that consider historical user data and each user's (individual's) preferences. The proposed algorithm was compared with a naive alternative using real-life Twitter dataset. The results showed that the proposed algorithm outperforms the naive alternative. (C) 2017 Elsevier Ltd. All rights reserved.Öğe Discovery of hydrometeorological patterns(TUBITAK SCIENTIFIC & TECHNICAL RESEARCH COUNCIL TURKEY, 2014) Celik, Mete; Dadaser-Celik, Filiz; Dokuz, Ahmet SakirHydrometeorological patterns can be defined as meaningful and nontrivial associations between hydrological and meteorological parameters over a region. Discovering hydrometeorological patterns is important for many applications, including forecasting hydrometeorological hazards (floods and droughts), predicting the hydrological responses of ungauged basins, and filling in missing hydrological or meteorological records. However, discovering these patterns is challenging due to the special characteristics of hydrological and meteorological data, and is computationally complex due to the archival history of the datasets. Moreover, defining monotonic interest measures to quantify these patterns is difficult. In this study, we propose a new monotonic interest measure, called the hydrometeorological prevalence index, and a novel algorithm for mining hydrometeorological patterns (HMP-Miner) out of large hydrological and meteorological datasets. Experimental evaluations using real datasets show that our proposed algorithm outperforms the naive alternative in discovering hydrometeorological patterns efficiently.Öğe Experimental and Numerical Investigation of the Effect on the Water Surface Profiles of Bridge Pier Type and Skew Angle(Konya Teknik Univ, 2023) Erduran, Kutsi Savas; Unal, Ugur; Dokuz, Ahmet Sakir; Nas, Mustafa CagriDesign of bridge piers is so much important for the structural safety of bridges, however, at least at the same extent, hydraulic analysis of bridge piers is also required. Bridge piers are structures that obstruct the flow by narrowing the flow area and can cause flooding in the bridge upstream. Therefore, with the light of numerical and experimental studies, accurate estimations of the occurrence of water surface profiles and affluxes resulted from the construction of bridge piers under different flow conditions are the major part of bridge hydraulics. In this study, the effects of bridge piers with four different cross-sections and four different skew angles on the water surface profiles and the affluxes were experimentally and numerically investigated. The experimental study has been conducted in a flume with dimensions of 10x0.309x0.45m in Department of Civil Engineering Hydraulics Laboratory at Nigde Omer Halisdemir University. The experimental measurements were taken using 4 different shapes of bridge piers (square, circular, oblong and ogival upstream face) and 4 different skew angles (0(0), 15(0), 30(0) and 45(0)), and the images obtained from video recordings of each experiment were converted into numerical values using an image processing technique. In addition, commercial package program HEC-RAS software and a direct step method were also used for the numerical modelling and the numerical water surface profiles and affluxes were also obtained for the purpose of comparisons. In the conclusion part, the comments on the applicability of HEC-RAS software and the suitability of the image processing technique were given for the similar problems.Öğe Fast and efficient discovery of key bike stations in bike sharing systems big datasets(Pergamon-Elsevier Science Ltd, 2021) Dokuz, Ahmet SakirBike Sharing Systems (BSS) became one of the popular transportation systems due to their environmental friendly, mobility endorsing, and outdoor activity nature. City residents tend to use particular bike stations more frequently and prevalent than other stations for different reasons, such as popularity and centrality of the lo-cations. Discovering key bike stations is the exploration of frequent and mostly utilized bike stations which are highly preferred by BSS users in terms of spatial and temporal activities. Discovering key stations is important for optimal planning for BSS, bike repositioning methods, and urban land use applications. However, discovering key stations is challenging due to variability of bike user preferences, effect of weather conditions, and big data nature of BSS datasets. In this study, two interest measures are proposed to discover key bike stations using BSS big datasets. Proposed interest measures reveal frequency and prevalence of stations in terms of daily and dataset-wide usage of BSS users. Two algorithms are proposed to discover key stations using proposed interest measures. One of the proposed algorithms could better handle BSS big datasets within less execution time and more efficient memory usage by dividing and processing each section of dataset separately. The proposed al-gorithms are experimentally evaluated using Chicago Divvy Bikes dataset. The results show that the proposed algorithms are effective on discovering key stations among BSS big datasets, which could be beneficial for making decisions about city resident mobility behaviors in terms of bike user activities and for extracting knowledge about bike user preferences.Öğe Investigating lane-changing moves of vehicles departing from signalized junction(Taylor & Francis Ltd, 2022) Bagdatli, Muhammed Emin Cihangir; Dokuz, Ahmet Sakir; Honul, AyetullahDiscretionary Lane-Changing (DLC) possesses different characteristics under various traffic conditions. Therefore, in order to model DLC moves in the correct manner, they need to be approached separately for various conditions. This study has examined the DLC moves of vehicles, which depart from signalized junction. It is observed that a large number of vehicles move to a neighboring lane in order to travel under better traffic conditions, after the traffic lights turn green. There are many parameters, which motivate the vehicles to change lanes at this stage. This study has focused on discovering these parameters. Data was collected through field work for this purpose. Thereinafter, the impact value of each parameter has been discovered using four separate feature selection methods and has been sorted. Fuzzy Cognitive Maps have been developed with the obtained impact values. The models have been validated using the field data and have been compared with each other.Öğe Location-based optimal sizing of hybrid renewable energy systems using deterministic and heuristic algorithms(Wiley, 2021) Demolli, Halil; Dokuz, Ahmet Sakir; Ecemis, Alper; Gokcek, MuratThe application of renewable energy sources in electrical energy generation is becoming widespread due to the decrease of installation costs and the increase of environmental concerns. Hybrid power generation systems are advantageous to meet the load demand, but optimal sizing is the main concern for having a cost-effective system based on given load demand and techno-economic indicators. This paper proposes a deterministic algorithm and utilizes genetic and artificial bee colony (ABC) optimization algorithms for optimal sizing of PV/battery and PV/WT/battery hybrid systems with minimum levelized cost of electricity (LCOE) constraint for two locations, Nigde and Bozcaada, in Turkey. The loss of power supply probability (LPSP) is used to build a reliable system and to make sure that the system produces required energy. Experimental results showed that optimal sizing of each location is different due to different wind and solar characteristics of locations. PV/battery model is more suitable for Nigde location with 1.22% LPSP and 0.1514 [$/kWh] LCOE, while PV/WT/battery model is more cost-efficient for Bozcaada location with 1.952% LPSP and 0.0872 [$/kWh] LCOE. Time performances of the algorithms are also investigated. It has been seen that the ABC algorithm has better performance and less execution time. This study demonstrated that heuristic algorithms are more applicable than deterministic algorithms, due to fast discovery of optimal solutions for hybrid renewable energy systems.Öğe Modeling discretionary lane-changing decisions using an improved fuzzy cognitive map with association rule mining(Taylor & Francis Ltd, 2021) Bagdatli, Muhammed Emin Cihangir; Dokuz, Ahmet SakirThe discretionary lane-changing process consists of two phases. The first phase is decision making on lane-changing, and the second phase is the execution of this decision. The first phase has a complex structure that is affected by many parameters. In this phase, some parameters are present that affect lane-changing directly, while some other indirect parameters motivate drivers to perform lane-changing. This study focuses on discovering the parameters that prompt drivers to change lanes. The parameters determined as a result of the interviews with the drivers were examined in the field study. Then, the impact of the parameters for lane-changing were discovered using association rule mining and the proposed Significant Association Features Extractor (SigAFE) algorithm. Fuzzy Cognitive Map (FCM) discretionary lane-changing decision models were developed using the impact values that were discovered using the SigAFE algorithm. The performances of the models were revealed with the actual data of the field study.Öğe Optimum sizing of hybrid renewable power systems for on-site hydrogen refuelling stations: Case studies from Türkiye and Spain(Pergamon-Elsevier Science Ltd, 2024) Gokcek, Murat; Paltrinieri, Nicola; Liu, Yiliu; Badia, Eulalia; Dokuz, Ahmet Sakir; Erdogmus, Ayse; Urhan, Baki BarisOne of the main barriers to the adoption of fuel cell vehicles (FCEVs) is the limited availability of hydrogen refuelling stations (HRSs). The presence of these stations is crucial in facilitating the provision of fuel for FCEVs, which rely on hydrogen as a source of power generation. Renewable energy sources offer significant advantages for hydrogen production at these stations, as they are environmentally friendly and can reduce costs. In this study, it is provided a techno-economic analysis of an on -site hydrogen refuelling station powered by a hybrid renewable energy generation system using HOMER software in Nigde, Turkiye, and Zaragoza, Spain. Three different power system scenarios were evaluated to refuel 24 vehicles per day for each region throughout the year. The results of the analysis showed that the most optimal system architecture for Nigde was with a solar panel power generation system, with a levelized cost of hydrogen (LCOH) of 6.15 $/kg and the net present cost (NPC) of $6,832,393. The most optimal system architecture for Zaragoza was a wind turbine -photovoltaic panel power generation system, with an LCOH of 5.83 $/kg and NPC of $6,499,723. The annual amount of CO2 emissions avoided by using renewable resources in hydrogen production was calculated as 2,673,453 kg for Nigde and 2,366,573 kg for Zaragoza. The study also found that the cost of hydrogen production increases with decreasing the HRS production capacity. The use of renewable energy generation systems for hydrogen production will enable countries to achieve net -zero emission targets and reduce the need to import fossil fuels to meet energy demands in the transportation sector. The present study makes several contributions towards the achievement of the United Nations Sustainable Development Goals (3, 7, 11, and 13) and has the potential to facilitate the rapid adoption of FCEVs.Öğe Social velocity based spatio-temporal anomalous daily activity discovery of social media users(Springer, 2022) Dokuz, Ahmet SakirAnomalous daily activities are the activities that do not fit into normal daily behavior of social media users. Discovering anomalous daily activities is important for protecting social media users from harmful content and providing correct information about populated accounts, products, or hashtags. However, discovering anomalous daily activities is challenging due to hardness of detection of bot applications, complexity of anomalous activities, and the big data nature of social media datasets. In this study, a novel method that discovers anomalous daily activities with respect to spatio-temporal information of social media datasets is proposed. For this purpose, an interest measure, named as social velocity, is proposed to discover anomalous daily activities that is based on spatial distance and temporal difference of successive posts. Two novel algorithms are proposed that use proposed method and interest measure and experimentally evaluated on a real Twitter dataset. The experimental results show that proposed algorithms are successful for discovering anomalous activities of social media users with respect to spatio-temporal information.Öğe Time-persistent regions discovery of taxi trajectory big datasets based on regional spatio-temporal velocity(Elsevier, 2023) Dokuz, Yesim; Dokuz, Ahmet SakirWith the increasing number of residents and motor vehicles in urban areas, traffic-related problems have emerged. Traffic analysis and prediction systems provide in-formation about the city dynamics and traffic estimation on a regional basis. Several studies are performed for traffic analysis and prediction in urban datasets, including taxi trajectory datasets, however, these studies do not focus on regional traffic analysis and time-persistent regions discovery. Time-persistent regions refer to the regions that have a stable utilization and relatively stationary velocity in terms of traffic activities. In this study, a novel method is proposed to discover time-persistent regions based on regional daily velocity values using taxi trajectory big datasets. A new algorithm, namely Time-Persistent Regions Discovery algorithm (TPRD algorithm), is proposed based on the proposed method. The proposed TPRD algorithm is experimentally evaluated on TLC Taxi Trip Records big dataset of New York City and the results show that the proposed algorithm could discover time-persistent regions based on proposed interest measures and threshold values. & COPY; 2023 Elsevier B.V. All rights reserved.Öğe Vehicle Delay Estimation at Signalized Intersections Using Machine Learning Algorithms(Sage Publications Inc, 2021) Bagdatli, Muhammed Emin Cihangir; Dokuz, Ahmet SakirAccurate determination of average vehicle delays is significant for effective management of a signalized intersection. The vehicle delays can be determined by field studies, however, this approach is costly and time consuming. Analytical methods which are commonly utilized to estimate delay cannot generate accurate estimates, especially in oversaturated traffic flow conditions. Delay estimation models based on artificial intelligence have been presented in the literature in recent years to estimate the delay more accurately. However, the number of artificial/heuristic intelligence techniques utilized for vehicle delay estimation is limited in the literature. In this study, estimation models are developed using four different machine learning methods-support vector regression (SVR), random forest (RF), k nearest neighbor (kNN), and extreme gradient boosting (XGBoost)-that have not previously been applied in the literature for vehicle delay estimation at signalized intersections. The models were tested with data collected from 12 signalized intersections located in Ankara, the capital of Turkey, and the performance of the models was revealed. The models were furthermore compared with successful delay models from the literature. The developed models, in particular the RF and XGBoost models, showed high performance in estimating the delay at signalized intersections under different traffic conditions. The results indicate that the delay estimation models based on the RF and XGBoost techniques can significantly contribute to both the literature and practice.Öğe Weighted spatio-temporal taxi trajectory big data mining for regional traffic estimation(Elsevier, 2022) Dokuz, Ahmet SakirThe estimation of traffic conditions in cities is becoming essential to establish a sustainable transportation system and to help traffic management authorities plan the traffic of cities. Recently, taxi trajectory big datasets are being collected during taxi drivers are routing around the cities. Taxi trajectory datasets provide behavioral information about the city residents, urban flows of the taxi passengers, and infrastructure for traffic condition estimation. This study aims to estimate regional traffic velocity of New York City using New York taxi trajectory dataset. A new method is proposed that uses weighted spatio-temporal trajectory big data mining approach and scores each region of the cities in terms of traffic velocity. A new algorithm is proposed, namely Regional Traffic Velocity Estimation (RTVE) algorithm, which uses proposed regional spatio-temporal velocity estimation method and experimentally evaluated using New York taxi trajectory dataset. Experimental results show that each region in New York have different velocity and usage characteristics in terms of hourly and daily analyses. Also, borough-level analyses are performed that reveal knowledge about the boroughs of New York. The estimated regional traffic velocity of cities based on taxi trajectory datasets would provide a decision support system for decision-makers in terms of regional hourly and daily evaluation of cities with cost-free and widespread city traffic dataset. (C) 2021 Elsevier B.V. All rights reserved.Öğe Wind power forecasting based on daily wind speed data using machine learning algorithms(Pergamon-Elsevier Science Ltd, 2019) Demolli, Halil; Dokuz, Ahmet Sakir; Ecemis, Alper; Gokcek, MuratWind energy is a significant and eligible source that has the potential for producing energy in a continuous and sustainable manner among renewable energy sources. However, wind energy has several challenges, such as initial investment costs, the stationary property of wind plants, and the difficulty in finding wind-efficient energy areas. In this study, long-term wind power forecasting was performed based on daily wind speed data using five machine learning algorithms. We proposed a method based on machine learning algorithms to forecast wind power values efficiently. We conducted several case studies to reveal performances of machine learning algorithms. The results showed that machine learning algorithms could be used for forecasting long-term wind power values with respect to historical wind speed data. Furthermore, the results showed that machine learning-based models could be applied to a location different from model-trained locations. This study demonstrated that machine learning algorithms could be successfully used before the establishment of wind plants in an unknown geographical location whether it is logical by using the model of a base location.
