Bilgisayar Mühendisliği Bölümü

Permanent URI for this communityhttps://hdl.handle.net/20.500.12416/13

Browse

Now showing 1 - 20 of 21

Citation - WoS: 188
Citation - Scopus: 278
Adoption of E-Government Services in Turkey
(Pergamon-elsevier Science Ltd, 2017) Arifoglu, Ali; Tokdemir, Gul; Pacin, Yudum; Kurfali, Murathan
This research aims to investigate underlying factors that play role in citizens' decision to use e-government services in Turkey. UTAUT model which was enriched by introducing Trust of internet and Trust of government factors is used in the study. The model is evaluated through a survey conducted with Turkish citizens who are from different regions of the country. A total of 529 answers collected through purposive sampling and the responses were evaluated with the SEM (Structural Equation Modeling) technique. According to the results, Performance expectancy, Social influence, Facilitating conditions and Trust of Internet were found to have a positive effect on behavioral intention to use e-government services. Additionally, both Trust factors were found to have a positive influence on Performance expectancy of e-government services, a relation which, to our best knowledge, hasn't been tested before in e-government context. Effect of Effort expectancy and Trust of government were found insignificant on behavioral intention. We believe that the findings of this study will guide professionals and policy makers in improving and popularizing e-government services by revealing the citizen's priorities regarding e-government services in Turkey. (C) 2016 Elsevier Ltd. All rights reserved.
Analysis of transfer learning for deep neural network based plant classification models
(2019) Kaya, Aydın; Keçeli, Ali Seydi; Çatal, Çağatay; Yalıç, Hamdi Yalın; Temuçin, Hüseyin; Tekinerdoğan, Bedir
Plant species classification is crucial for biodiversity protection and conservation. Manual classification is time-consuming, expensive, and requires experienced experts who are often limited available. To cope with these issues, various machine learning algorithms have been proposed to support the automated classification of plant species. Among these machine learning algorithms, Deep Neural Networks (DNNs) have been applied to different data sets. DNNs have been however often applied in isolation and no effort has been made to reuse and transfer the knowledge of different applications of DNNs. Transfer learning in the context of machine learning implies the usage of the results of multiple applications of DNNs. In this article, the results of the effect of four different transfer learning models for deep neural network-based plant classification is investigated on four public datasets. Our experimental study demonstrates that transfer learning can provide important benefits for automated plant identification and can improve low-performance plant classification models.
Citation - WoS: 40
Citation - Scopus: 52
Automated Classification of Rheumatoid Arthritis, Osteoarthritis, and Normal Hand Radiographs With Deep Learning Methods
(Springer, 2022) Maras, Hadi Hakan; Ureten, Kemal
Rheumatoid arthritis and hand osteoarthritis are two different arthritis that causes pain, function limitation, and permanent joint damage in the hands. Plain hand radiographs are the most commonly used imaging methods for the diagnosis, differential diagnosis, and monitoring of rheumatoid arthritis and osteoarthritis. In this retrospective study, the You Only Look Once (YOLO) algorithm was used to obtain hand images from original radiographs without data loss, and classification was made by applying transfer learning with a pre-trained VGG-16 network. The data augmentation method was applied during training. The results of the study were evaluated with performance metrics such as accuracy, sensitivity, specificity, and precision calculated from the confusion matrix, and AUC (area under the ROC curve) calculated from ROC (receiver operating characteristic) curve. In the classification of rheumatoid arthritis and normal hand radiographs, 90.7%, 92.6%, 88.7%, 89.3%, and 0.97 accuracy, sensitivity, specificity, precision, and AUC results, respectively, and in the classification of osteoarthritis and normal hand radiographs, 90.8%, 91.4%, 90.2%, 91.4%, and 0.96 accuracy, sensitivity, specificity, precision, and AUC results were obtained, respectively. In the classification of rheumatoid arthritis, osteoarthritis, and normal hand radiographs, an 80.6% accuracy result was obtained. In this study, to develop an end-to-end computerized method, the YOLOv4 algorithm was used for object detection, and a pre-trained VGG-16 network was used for the classification of hand radiographs. This computer-aided diagnosis method can assist clinicians in interpreting hand radiographs, especially in rheumatoid arthritis and osteoarthritis.
Citation - WoS: 284
Citation - Scopus: 365
Context-Aware Computing, Learning, and Big Data in Internet of Things: a Survey
(Ieee-inst Electrical Electronics Engineers inc, 2018) Dogdu, Erdogan; Ozbayoglu, Ahmet Murat; Sezer, Omer Berat
Internet of Things (IoT) has been growing rapidly due to recent advancements in communications and sensor technologies. Meanwhile, with this revolutionary transformation, researchers, implementers, deployers, and users are faced with many challenges. IoT is a complicated, crowded, and complex field; there are various types of devices, protocols, communication channels, architectures, middleware, and more. Standardization efforts are plenty, and this chaos will continue for quite some time. What is clear, on the other hand, is that IoT deployments are increasing with accelerating speed, and this trend will not stop in the near future. As the field grows in numbers and heterogeneity, "intelligence" becomes a focal point in IoT. Since data now becomes "big data," understanding, learning, and reasoning with big data is paramount for the future success of IoT. One of the major problems in the path to intelligent IoT is understanding "context," or making sense of the environment, situation, or status using data from sensors, and then acting accordingly in autonomous ways. This is called "context-aware computing," and it now requires both sensing and, increasingly, learning, as IoT systems get more data and better learning from this big data. In this survey, we review the field, first, from a historical perspective, covering ubiquitous and pervasive computing, ambient intelligence, and wireless sensor networks, and then, move to context-aware computing studies. Finally, we review learning and big data studies related to IoT. We also identify the open issues and provide an insight for future study areas for IoT researchers.
Citation - WoS: 38
Citation - Scopus: 49
Development of a Recurrent Neural Networks-Based Calving Prediction Model Using Activity and Behavioral Data
(Elsevier Sci Ltd, 2020) Keceli, Ali Seydi; Catal, Cagatay; Kaya, Aydin; Tekinerdogan, Bedir
Accurate prediction of calving time in dairy cattle is crucial for dairy herd management to reduce risks like dystocia and pain. Prediction of calving using traditional, manual observation such as observing breeding records and visual cues, however, is a complicated and error-prone task whereby even experts can fail to provide a proper prediction. Moreover, manual prediction does not scale for larger farms and becomes very soon time-consuming, inefficient, and costly. In this context, automated solutions are considered to be promising to provide both better and more efficient predictions, thereby supporting the health of the dairy cows and reducing the unnecessary overhead for farmers. Although the first automated solutions appear to have mainly focused on statistical solutions, currently, machine learning approaches are now increasingly being considered as a feasible and promising approach for accurate prediction of calving. In this context, the objective of this study is to develop machine learning-based prediction models that provide higher performance compared to the existing tools, methods, and techniques. This study shows that the calving of the cattle can be predicted by applying several behaviors of cattle, behavioral monitoring sensors, and machine learning models. Bi-directional Long Short-Term Memory (Bi-LSTM) method has been applied for the prediction of the calving day, and the RusBoosted Tree classifier has been used to predict the remaining 8 h before calving. The experimental results demonstrated that Bi-LSTM provides better performance compared to the LSTM algorithm in terms of classification accuracy, while the RusBoosted Tree algorithm predicts the remaining 8 h accurately before calving. Furthermore, Recurrent Neural Networks provide high performance for the prediction of calving day.
Citation - WoS: 34
Citation - Scopus: 40
Diffusion of Latent Semantic Analysis as a Research Tool: a Social Network Analysis Approach
(Elsevier, 2010) Tonta, Yasar; Darvish, Hamid R.
Latent semantic analysis (LSA) is a relatively new research tool with a wide range of applications in different fields ranging from discourse analysis to cognitive science, from information retrieval to machine learning and so on. In this paper, we chart the development and diffusion of LSA as a research tool using social network analysis (SNA) approach that reveals the social structure of a discipline in terms of collaboration among scientists. Using Thomson Reuters' Web of Science (WoS), we identified 65 papers with "latent semantic analysis" in their titles and 250 papers in their topics (but not in titles) between 1990 and 2008. We then analyzed those papers using bibliometric and SNA techniques such as co-authorship and cluster analysis. It appears that as the emphasis moves from the research tool (LSA) itself to its applications in different fields, citations to papers with LSA in their titles tend to decrease. The productivity of authors fits Lotka's Law while the network of authors is quite loose. Networks of journals cited in papers with LSA in their titles and topics are well connected. (C) 2009 Elsevier Ltd. All rights reserved.
Citation - WoS: 71
Citation - Scopus: 100
An Examination of Personality Traits and How They Impact on Software Development Teams
(Elsevier, 2017) O'Connor, Rory V.; Colomo-Palacios, Ricardo; Clarke, Paul; Yilmaz, Murat
Context Research has shown that a significant number of software projects fail due to social issues such as team or personality conflicts. However, only a limited number of empirical studies have been undertaken to understand the impact of individuals' personalities on software team configurations. These studies suffer from an important limitation as they lack a systematic and rigorous method to relate personality traits of software practitioners and software team structures. Objective: Based on an interactive personality profiling approach, the goal of this study is to reveal the personality traits of software practitioners with an aim to explore effective software team structures. Method: To explore the importance of individuals' personalities on software teams, we employed a two-step empirical approach. Firstly, to assess the personality traits of software practitioners, we developed a context-specific survey instrument, which was conducted on 216 participants from a middle-sized soft ware company. Secondly, we propose a novel team personality illustration method to visualize team structures. Results: Study results indicated that effective team structures support teams with higher emotional stability, agreeableness, extroversion, and conscientiousness personality traits. Conclusion: Furthermore, empirical results of the current study show that extroversion trait was more predominant than previously suggested in the literature, which was especially more observable among agile software development teams. (C) 2017 Elsevier B.V. All rights reserved.
Citation - WoS: 4
Citation - Scopus: 5
Feynman, Biominerals and Graphene - Basic Aspects of Nanoscience
(Elsevier, 2010) Ozdogan, Cem; Quandt, Alexander
This article is about writing small. Inspired by R.P. Feynman's legendary talk There's plenty of room at the bottom, we recapitulate his famous Gedanken experiment of condensing a lot of useful information on the head of a pin [see Feymnan R, J. MEMS 1 (1992) 60]. These considerations will familiarize LIS with the length scales for a future downsizing of technological components, and they allow for some speculations about ultimate physical or chemical limits of the corresponding nanodevices. Furthermore we will analyze the nano-technological capabilities of Mother Nature in the case of magnetotactic bacteria, and briefly sketch the cornerstones of the rapidly growing field of biomineralization, which might open up a new science of complex functional nanomaterials in the near future. Finally we describe a general scheme to shrink integrated microelectronic circuits towards the very size limits of nanotechnology. (C) 2009 Elsevier B.V. All rights reserved.
Citation - WoS: 111
Citation - Scopus: 137
Hybrid Expert Systems: a Survey of Current Approaches and Applications
(Pergamon-elsevier Science Ltd, 2012) Tolun, M. R.; Hassanpour, R.; Sahin, S.
This paper is a statistical analysis of hybrid expert system approaches and their applications but more specifically connectionist and neuro-fuzzy system oriented articles are considered. The current survey of hybrid expert systems is based on the classification of articles from 1988 to 2010. Present analysis includes 91 articles from related academic journals, conference proceedings and literature reviews. Our results show an increase in the number of recent publications which is an indication of gaining popularity on the part of hybrid expert systems. This increase in the articles is mainly in neuro-fuzzy and rough neural expert systems' areas. We also observe that many new industrial applications are developed using hybrid expert systems recently. (C) 2011 Elsevier Ltd. All rights reserved.
Citation - WoS: 16
Citation - Scopus: 18
The Impact of Incapacitation of Multiple Critical Sensor Nodes on Wireless Sensor Network Lifetime
(Ieee-inst Electrical Electronics Engineers inc, 2017) Tavli, Bulent; Kahjogh, Behnam Ojaghi; Dogdu, Erdogan; Yildiz, Huseyin Ugur
Wireless sensor networks (WSNs) are envisioned to be utilized in many application areas, such as critical infrastructure monitoring, and therefore, WSN nodes are potential targets for adversaries. Network lifetime is one of the most important performance indicators in WSNs. The possibility of reducing the network lifetime significantly by eliminating a certain subset of nodes through various attacks will create the opportunity for the adversaries to hamper the performance of WSNs with a low risk of detection. However, the extent of reduction in network lifetime due to elimination of a group of critical sensor nodes has never been investigated in the literature. Therefore, in this letter, we create two novel algorithms based on a linear programming framework to model and analyze the impact of critical node elimination attacks on WSNs and explore the parameter space through numerical evaluations of the algorithms. Our results show that critical node elimination attacks can significantly shorten the network lifetime.
Citation - WoS: 2
Citation - Scopus: 3
Mining Medline for the Treatment of Osteoporosis
(Springer, 2012) Ceken, Cinar; Hassanpour, Reza; Esmelioglu, Sadik; Tolun, Mehmet Resit; Yildirim, Pinar
In this paper, we consider the importance of osteoporosis disease in terms of medical research and pharmaceutical industry and we introduce a knowledge discovery approach regarding the treatment of osteoporosis from a historical perspective. Osteoporosis is a systemic skeletal disease in which osteoporotic fractures are associated with substantial morbidity and mortality and impaired quality of life. Osteoporosis has also higher costs, for example, longer hospital stays than many other diseases such as diabetes and heart attack and it is an attractive market for pharmaceutical companies. We use a freely available biomedical search engine leveraging text-mining technology to extract the drug names used in the treatment of osteoporosis from MEDLINE articles. We conclude that alendronate (Fosamax) and raloxifene (Evista) have the highest number of articles in MEDLINE and seem the dominating drugs for the treatment of osteoporosis in the last decade.
Citation - WoS: 42
Citation - Scopus: 50
Mobile Language Learning: Contribution of Multimedia Messages Via Mobile Phones in Consolidating Vocabulary
(Springer Heidelberg, 2012) Saran, Murat; Saran, Murat; Seferoglu, Golge; Cagiltay, Kursat; Bilgisayar Mühendisliği
This study aimed at investigating the effectiveness of using multimedia messages via mobile phones in helping language learners in consolidating vocabulary. The study followed a pre-test/post-test quasi-experimental research design. The participants of this study were a group of students attending the English Preparatory School of an English-medium university in Turkey. Six different groups were formed in order to investigate the comparative effectiveness of supplementary vocabulary materials delivered through three different means: via mobile phones, on web pages, and in print form. The multimedia messages in this study included the definitions of words, exemplary sentences, related visual representations, information on word formation, and pronunciations of words. Analyses of the quantitative data showed that using mobile phones had positive effects on students' vocabulary acquisition. The results suggest that mobile phones offer great potential for providing learners with supplementary opportunities to recontextualize, recycle, and consolidate vocabulary.
Citation - WoS: 13
Citation - Scopus: 19
New Knowledge in Strategic Management Through Visually Mining Semantic Networks
(Springer, 2017) Tokdemir, Gul; Sevinc, Mete; Tunc, Murat Mustafa; Ertek, Gurdal
Today's highly competitive business world requires that managers be able to make fast and accurate strategic decisions, as well as learn to adapt to new strategic challenges. This necessity calls for a deep experience and a dynamic understanding of strategic management. The trait of dynamic understanding is mainly the skill of generating additional knowledge and innovative solutions under the new environmental conditions. Building on the concepts of information processing, this paper aims to support managers in constructing new strategic management knowledge, through representing and mining existing knowledge through graph visualization. To this end, a three-stage framework is proposed and described. The framework can enable managers to develop a deeper understanding of the strategic management domain, and expand on existing knowledge through visual analysis. The model further supports a case study that involves unstructured knowledge of profit patterns and the related strategies to succeed using these patterns. The applicability of the framework is shown in the case study, where the unstructured knowledge in a strategic management book is first represented as a semantic network, and then visually mined for revealing new knowledge.
Citation - WoS: 2
Citation - Scopus: 3
A New Relational Learning System Using Novel Rule Selection Strategies
(Elsevier, 2006) Uludag, Mahmut; Tolun, Mehmet R.
This paper describes a new rule induction system, rila, which can extract frequent patterns from multiple connected relations. The system supports two different rule selection strategies, namely the select early and select late strategies. Pruning heuristics are used to control the number of hypotheses generated during the learning process. Experimental results are provided on the mutagenesis and the segmentation data sets. The present rule induction algorithm is also compared to the similar relational learning algorithms. Results show that the algorithm is comparable to similar algorithms. (c) 2006 Elsevier B.V. All rights reserved.
Citation - WoS: 6
Citation - Scopus: 16
Parallel Wavecluster: a Linear Scaling Parallel Clustering Algorithm Implementation With Application To Very Large Datasets
(Academic Press inc Elsevier Science, 2011) Ozdogan, Cem; Yildirim, Ahmet Artu
A linear scaling parallel clustering algorithm implementation and its application to very large datasets for cluster analysis is reported. WaveCluster is a novel clustering approach based on wavelet transforms. Despite this approach has an ability to detect clusters of arbitrary shapes in an efficient way, it requires considerable amount of time to collect results for large sizes of multi-dimensional datasets. We propose the parallel implementation of the WaveCluster algorithm based on the message passing model for a distributed-memory multiprocessor system. In the proposed method, communication among processors and memory requirements are kept at minimum to achieve high efficiency. We have conducted the experiments on a dense dataset and a sparse dataset to measure the algorithm behavior appropriately. Our results obtained from performed experiments demonstrate that developed parallel WaveCluster algorithm exposes high speedup and scales linearly with the increasing number of processors. (C) 2011 Elsevier Inc. All rights reserved.
Citation - WoS: 22
Citation - Scopus: 37
Performing and Analyzing Non-Formal Inspections of Entity Relationship Diagram (Erd)
(Elsevier Science inc, 2013) Tokdemir, Gul; Kilic, Ozkan; Topalli, Damla; Cagiltay, Nergiz Ercil
Designing and understanding of diagrammatic representations is a critical issue for the success of software projects because diagrams in this field provide a collection of related information with various perceptual signs and they help software engineers to understand operational systems at different levels of information system development process. Entity relationship diagram (ERD) is one of the main diagrammatic representations of a conceptual data model that reflects users' data requirements in a database system. In today's business environment, the business model is in a constant change which creates highly dynamic data requirements which also requires additional processes like modifications of ERD. However, in the literature there are not many measures to better understand the behaviors of software engineers during designing and understanding these representations. Hence, the main motivation of this study is to develop measures to better understand performance of software engineers during their understanding process of ERD. Accordingly, this study proposes two measures for ERD defect detection process. The defect detection difficulty level (DF) measures how difficult a defect to be detected according to the other defects for a group of software engineers. Defect detection performance (PP) measure is also proposed to understand the performance of a software engineer during the defect detection process. The results of this study are validated through the eye tracker data collected during the defect detection process of participants. Additionally, a relationship between the defect detection performance (PP) of a software engineer and his/her search patterns within an ERD is analyzed. Second experiment with five participants is also conducted to show the correlation between the proposed metric results and eye tracker data. The results of experiment-2 also found to be similar for DF and PP values. The results of this study are expected to provide insights to the researchers, software companies, and to the educators to improve ERD reasoning process. Through these measures several design guidelines can be developed for better graphical representations and modeling of the information which would improve quality of these diagrams. Moreover, some reviewing instructions can be developed for the software engineers to improve their reviewing process in ERD. These guidelines in turn will provide some tools for the educators to improve design and review skills of future software engineers. (c) 2013 Elsevier Inc. All rights reserved.
Citation - WoS: 7
Citation - Scopus: 10
Prediction of Similarities Among Rheumatic Diseases
(Springer, 2012) Ceken, Cinar; Hassanpour, Reza; Tolun, Mehmet Resit; Yildirim, Pinar
We introduce a method for extracting hidden patterns seen in rheumatic diseases by using articles from the widely used biomedical database MEDLINE. Rheumatic diseases affect hundreds of millions of people worldwide and lead to substantial loss of functioning and mobility. Diagnosing rheumatic diseases can be difficult because some symptoms are common to many of them. We use Facta system as a biomedical text mining tool for finding symptoms and then create a dataset with the frequencies of symptoms for each disease and apply hierarchical clustering analysis to find similarities between diseases. Clustering analysis yields four distinct types or groups of rheumatic diseases. Although our results cannot remove all the uncertainty for the diagnosis of rheumatic diseases, we believe they can contribute to the diagnosis of rheumatic diseases to a certain extent. We hope that some similarities exposed can provide additional information at the stage of decision-making.
Citation - WoS: 21
Citation - Scopus: 27
Reporting and Analyzing Alternative Clustering Solutions by Employing Multi-Objective Genetic Algorithm and Conducting Experiments on Cancer Data
(Elsevier, 2014) Peng, Peter; Addam, Omer; Ozyer, Sibel T.; Elzohbi, Mohamad; Elhajj, Ahmad; Gao, Shang; Alhajj, Reda
Clustering is an essential research problem which has received considerable attention in the research community for decades. It is a challenge because there is no unique solution that fits all problems and satisfies all applications. We target to get the most appropriate clustering solution for a given application domain. In other words, clustering algorithms in general need prior specification of the number of clusters, and this is hard even for domain experts to estimate especially in a dynamic environment where the data changes and/or become available incrementally. In this paper, we described and analyze the effectiveness of a robust clustering algorithm which integrates multi-objective genetic algorithm into a framework capable of producing alternative clustering solutions; it is called Multi-objective K-Means Genetic Algorithm (MOKGA). We investigate its application for clustering a variety of datasets, including microarray gene expression data. The reported results are promising. Though we concentrate on gene expression and mostly cancer data, the proposed approach is general enough and works equally to cluster other datasets as demonstrated by the two datasets Iris and Ruspini. After running MOKGA, a pareto-optimal front is obtained, and gives the optimal number of clusters as a solution set. The achieved clustering results are then analyzed and validated under several cluster validity techniques proposed in the literature. As a result, the optimal clusters are ranked for each validity index. We apply majority voting to decide on the most appropriate set of validity indexes applicable to every tested dataset. The proposed clustering approach is tested by conducting experiments using seven well cited benchmark data sets. The obtained results are compared with those reported in the literature to demonstrate the applicability and effectiveness of the proposed approach. (C) 2013 Elsevier B.V. All rights reserved.
Citation - WoS: 17
Citation - Scopus: 26
Scalable and Accurate Graph Clustering and Community Structure Detection
(Ieee Computer Soc, 2013) Onus, Melih; Djidjev, Hristo N.
One of the most useful measures of cluster quality is the modularity of the partition, which measures the difference between the number of the edges joining vertices from the same cluster and the expected number of such edges in a random graph. In this paper, we show that the problem of finding a partition maximizing the modularity of a given graph G can be reduced to a minimum weighted cut (MWC) problem on a complete graph with the same vertices as G. We then show that the resulting minimum cut problem can be efficiently solved by adapting existing graph partitioning techniques. Our algorithm finds clusterings of a comparable quality and is much faster than the existing clustering algorithms.
Citation - WoS: 17
Citation - Scopus: 24
Software Professionals During the Covid-19 Pandemic in Turkey: Factors Affecting Their Mental Well-Being and Work Engagement in the Home-Based Work Setting
(Elsevier Science inc, 2022) Tokdemir, Gul
With the COVID-19 pandemic, strict measures have been taken to slow down the spread of the virus, and consequently, software professionals have been forced to work from home. However, home based working entails many challenges, as the home environment is shared by the whole family simultaneously under pandemic conditions. The aim of this study is to explore software professionals' mental well-being and work engagement and the relationships of these variables with job strain and resource-related factors in the forced home-based work setting during the COVID-19 pandemic. An online cross-sectional survey based on primarily well-known, validated scales was conducted with software professionals in Turkey. The analysis of the results was performed through hierarchical multivariate regression. The results suggest that despite the negative effect of job strain, the resource related protective factors, namely, sleep quality, decision latitude, work-life balance, exercise predict mental well-being. Additionally, work engagement is predicted by job strain, sleep quality, and decision latitude. The results of the study will provide valuable insights to management of the software companies and professionals about the precautions that can be taken to have a better home-based working experience such as allowing greater autonomy and enhancing the quality of sleep and hence mitigating the negative effects of pandemic emergency situations on software professionals' mental well-being and work engagement. (C)& nbsp;2022 Elsevier Inc. All rights reserved.

Browse

Browsing Bilgisayar Mühendisliği Bölümü by WoS Q "Q1"