Yazılım Mühendisliği Bölümü Yayın Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/2147
Browse
13 results
Search Results
Article Citation - WoS: 38Citation - Scopus: 44Hyper-Heuristics: a Survey and Taxonomy(Pergamon-elsevier Science Ltd, 2024) Kucukyilmaz, Tayfun; Talbi, El-Ghazali; Dokeroglu, TanselHyper-heuristics are search techniques for selecting, generating, and sequencing (meta)-heuristics to solve challenging optimization problems. They differ from traditional (meta)-heuristics methods, which primarily employ search space-based optimization strategies. Due to the remarkable performance of hyper-heuristics in multi-objective and machine learning-based optimization, there has been an increasing interest in this field. With a fresh perspective, our work extends the current taxonomy and presents an overview of the most significant hyper-heuristic studies of the last two decades. Four categories under which we analyze hyperheuristics are selection hyper-heuristics (including machine learning techniques), low-level heuristics, target optimization problems, and parallel hyper-heuristics. Future research prospects, trends, and prospective fields of study are also explored.Article Citation - WoS: 2Citation - Scopus: 4Exploring Mooc Learners' Behavioural Patterns Considering Age, Gender and Number of Course Enrolments: Insights for Improving Educational Opportunities(int Council Open & Distance Education, 2024) Cagiltay, Nergiz ercil; Toker, Sacip; Cagiltay, KursatMassive Open Online Courses (MOOCs) now offer a variety of options for everyone to obtain a high -quality education. The purpose of this study is to better understand the behaviours of MOOC learners and provide some insights for taking actions that benefit larger learner groups. Accordingly, 2,288,559 learners' behaviours on 174 MITx courses were analysed. The results show that MOOCs are more attractive to the elderly, male, and highly educated groups of learners. Learners' performance improves as they register for more courses and improve their skills and experiences on MOOCs. The findings suggest that, in the long run, learners' adaptation to MOOCs will significantly improve the potential benefits of the MOOCs. Hence, MOOCs should continue by better understanding their learners and providing alternative instructional designs by considering different learner groups. MOOC providers' decision -makers may take these findings into account when making operational decisions.Article Citation - WoS: 5Citation - Scopus: 7Ranking Surgical Skills Using an Attention-Enhanced Siamese Network With Piecewise Aggregated Kinematic Data(Springer Heidelberg, 2022) Gilgien, Matthias; Ozdemir, Suat; Ogul, Burcin BuketPurpose Surgical skill assessment using computerized methods is considered to be a promising direction in objective performance evaluation and expert training. In a typical architecture for computerized skill assessment, a classification system is asked to assign a query action to a predefined category that determines the surgical skill level. Since such systems are still trained by manual, potentially inconsistent annotations, an attempt to categorize the skill level can be biased by potentially scarce or skew training data. Methods We approach the skill assessment problem as a pairwise ranking task where we compare two input actions to identify better surgical performance. We propose a model that takes two kinematic motion data acquired from robot-assisted surgery sensors and report the probability of a query sample having a better skill than a reference one. The model is an attention-enhanced Siamese Long Short-Term Memory Network fed by piecewise aggregate approximation of kinematic data. Results The proposed model can achieve higher accuracy than existing models for pairwise ranking in a common dataset. It can also outperform existing regression models when applied in their experimental setup. The model is further shown to be accurate in individual progress monitoring with a new dataset, which will serve as a strong baseline. Conclusion This relative assessment approach may overcome the limitations of having consistent annotations to define skill levels and provide a more interpretable means for objective skill assessment. Moreover, the model allows monitoring the skill development of individuals by comparing two activities at different time points.Article Citation - WoS: 5Citation - Scopus: 8Predicting the Severity of Covid-19 Patients Using a Multi-Threaded Evolutionary Feature Selection Algorithm(Wiley, 2022) Kiziloz, Hakan Ezgi; Sevinc, Ender; Dokeroglu, Tansel; Deniz, AycaThe COVID-19 pandemic has huge effects on the global community and an extreme burden on health systems. There are more than 185 million confirmed cases and 4 million deaths as of July 2021. Besides, the exponential rise in COVID-19 cases requires a quick prediction of the patients' severity for better treatment. In this study, we propose a Multi-threaded Genetic feature selection algorithm combined with Extreme Learning Machines (MG-ELM) to predict the severity level of the COVID-19 patients. We conduct a set of experiments on a recently published real-world dataset. We reprocess the dataset via feature construction to improve the learning performance of the algorithm. Upon comprehensive experiments, we report the most impactful features and symptoms for predicting the patients' severity level. Moreover, we investigate the effects of multi-threaded implementation with statistical analysis. In order to verify the efficiency of MG-ELM, we compare our results with traditional and state-of-the-art techniques. The proposed algorithm outperforms other algorithms in terms of prediction accuracy.Article Citation - WoS: 2Citation - Scopus: 4An Intelligent System for Detecting Mediterranean Fruit Fly [Medfly; Ceratitis Capitata (Wiedemann)](Pagepress Publ, 2022) Eyyuboglu, Halil Tanyer; Sari, Filiz; Uzun, Yusuf; Tolun, Mehmet ResitNowadays, the most critical agriculture-related problem is the harm caused to fruit, vegetable, nut, and flower crops by harmful pests, particularly the Mediterranean fruit fly, Ceratitis capitata, named Medfly. Medfly's existence in agricultural fields must be monitored systematically for effective combat against it. Special traps are utilised in the field to catch Medflies which will reveal their presence and applying pesticides at the right time will help reduce their population. A technologically supported automated remote monitoring system should eliminate frequent site visits as a more economical solution. This paper develops a deep learning system that can detect Medfly images on a picture and count their numbers. A particular trap equipped with an integrated camera that can take photos of the sticky band where Medflies are caught daily is utilised. Obtained pictures are then transmitted by an electronic circuit containing a SIM card to the central server where the object detection algorithm runs. This study employs a faster region-based convolutional neural network (Faster R-CNN) model in identifying trapped Medflies. When Medflies or other insects stick on the trap's sticky band, they spend extraordinary effort trying to release themselves in a panic until they die. Therefore, their shape is badly distorted as their bodies, wings, and legs are buckled. The challenge is that the deep learning system should detect these Medflies of distorted shape with high accuracy. Therefore, it is crucial to utilise pictures containing trapped Medfly images with distorted shapes for training and validation. In this paper, the success rate in identifying Medflies when other insects are also present is approximately 94%, achieved by the deep learning system training process, owing to the considerable amount of purpose-specific photographic data. This rate may be seen as quite favourable when compared to the success rates provided in the literature.Article Citation - WoS: 238Citation - Scopus: 308A Comprehensive Survey on Recent Metaheuristics for Feature Selection(Elsevier, 2022) Dokeroglu, Tansel; Deniz, Ayca; Kiziloz, Hakan EzgiFeature selection has become an indispensable machine learning process for data preprocessing due to the ever-increasing sizes in actual data. There have been many solution methods proposed for feature selection since the 1970s. For the last two decades, we have witnessed the superiority of metaheuristic feature selection algorithms, and tens of new ones are being proposed every year. This survey focuses on the most outstanding recent metaheuristic feature selection algorithms of the last two decades in terms of their performance in exploration/exploitation operators, selection methods, transfer functions, fitness value evaluations, and parameter setting techniques. Current challenges of the metaheuristic feature selection algorithms and possible future research topics are examined and brought to the attention of the researchers as well.Article Citation - WoS: 8Citation - Scopus: 7Assessment of Improvement of the Iri Model for Fof2 Variability Over Three Latitudes in Different Hemispheres During Low and High Solar Activities(Pergamon-elsevier Science Ltd, 2021) Timocin, Erdinc; Temucin, Huseyin; Inyurt, Samed; Shah, Munawar; Jamjareegulgarn, PunyawiThis paper discusses the diurnal and seasonal variations of the F2 layer critical frequency (foF2) and the improvement of performance of the IRI-2016 model in predicting foF2 over three latitudes in different hemispheres during low and high solar activities. We extracted the foF2 data from six ionosonde stations which are Manila (14.7 degrees N, 121.1 degrees E), Yamagawa (31.2 degrees N, 130.6 degrees E), Yakutsk (62.0 degrees N,129.6 degrees E), Townsville (19.6 degrees S, 146.8 degrees E), Hobart (42.9 degrees S, 147.3 degrees E) and Terre Adelie (66.6 degrees S, 140.0 degrees E). The data of both low solar activity (LSA) period and high solar activity (HSA) periods were divided into three seasons as Northern Summer (May, June, July and August), Equinoxes (March, April, September and October) and Northern Winter (November, December, January and February). The present study showed that the IRI-2016 performance is strongly dependent on the solar activity, latitude, season, local time and hemisphere. For both hemispheres, the foF2 values at low latitude station are larger than those at middle latitude station, whereas the foF2 values at middle latitude station are larger than those at high latitude station. The agreement between IRI2016-modelled foF2 and foF2 measurements on all stations selected in the northern hemisphere is best for North Summer and worst for North Winter. For northern hemisphere, the values of relative deviations during both solar activities are largest in high latitudes and smallest in middle latitudes. As for southern hemisphere, the values of relative deviations during LSA are largest in middle latitudes and smallest in high latitudes, whereas the values of relative deviations during HSA are largest in low latitudes and smallest in high latitudes. It is thought that the relative deviations in the observed foF2 values are caused by solar activity that strongly alter chemical and electromagnetic processes in the ionosphere. These results are important for future improvements depending on solar activity and seasons in the IRI model for foF2 values over three latitudes in different hemispheres.Article Citation - WoS: 7Citation - Scopus: 13A Concept-Based Sentiment Analysis Approach for Arabic(Zarka Private Univ, 2020) Sever, Hayri; Nasser, AhmedConcept-Based Sentiment Analysis (CBSA) methods are considered to be more advanced and more accurate when it compared to ordinary Sentiment Analysis methods, because it has the ability of detecting the emotions that conveyed by multi-word expressions concepts in language. This paper presented a CBSA system for Arabic language which utilizes both of machine learning approaches and concept-based sentiment lexicon. For extracting concepts from Arabic, a rule-based concept extraction algorithm called semantic parser is proposed. Different types of feature extraction and representation techniques are experimented among the building prosses of the sentiment analysis model for the presented Arabic CBSA system. A comprehensive and comparative experiments using different types of classification methods and classifier fusion models, together with different combinations of our proposed feature sets, are used to evaluate and test the presented CBSA system. The experiment results showed that the best performance for the sentiment analysis model is achieved by combined Support Vector Machine-Logistic Regression (SVM-LR) model where it obtained a F-score value of 93.23% using the Concept-Based-Features + Lexicon-Based-Features + Word2vec-Features (CBF + LEX+ W2V) features combinations.Article A Two-Stage Matching Method for Multi-Component Shapes(Univ Suceava, Fac Electrical Eng, 2015) Hassanpour, RezaIn this paper a shape matching algorithm for multiple component objects has been proposed which aims at matching shapes by a two-stage method. The first stage extracts the similarity features of each component using a generic shape representation model. The first stage of our shape matching method normalizes the components for orientation and scaling, and neglects minor deformations. In the second stage, the extracted similarity features of the components are combined with their relative spatial characteristics for shape matching. Some important application areas for the proposed multi-component shape matching are medical image registration, content based medical image retrieval systems, and matching articulated objects which rely on the a-priori information of the model being searched. In these applications, salient features such as vertebrae or rib cage bones can be easily segmented and used. These features however, show differences from person to person on one hand and similarities at different cross-sectional images of the same examination on the other hand. The proposed method has been tested on articulated objects, and reliable registration of 3-dimensional abdominal computed tomography images.Article Citation - WoS: 1Citation - Scopus: 3Identifying Criminal Organizations From Their Social Network Structures(Tubitak Scientific & Technological Research Council Turkey, 2019) Genc, Burkay; Sever, Hayri; Cinar, Muhammet SerkanIdentification of criminal structures within very large social networks is an essential security feat. By identifying such structures, it may be possible to track, neutralize, and terminate the corresponding criminal organizations before they act. We evaluate the effectiveness of three different methods for classifying an unknown network as terrorist, cocaine, or noncriminal. We consider three methods for the identification of network types: evaluating common social network analysis metrics, modeling with a decision tree, and network motif frequency analysis. The empirical results show that these three methods can provide significant improvements in distinguishing all three network types. We show that these methods are viable enough to be used as supporting evidence by security forces in their fight against criminal organizations operating on social networks.