Computer Engineering / Bilgisayar Mühendisliği

Permanent URI for this collectionhttps://hdl.handle.net/11147/10

Browse

Filters

2005 - 200932010 - 201932020 - 20233

Settings

Has files: NoWoS Q: search.filters.wosQuartile.N/A

Search Results

Now showing 1 - 9 of 9
  • Conference Object
    Citation - Scopus: 11
    An Analysis of Large Language Models and Langchain in Mathematics Education
    (Association for Computing Machinery, 2023) Soygazi,F.; Oğuz, Damla
    The development of large language models (LLMs) has led to the consideration of new approaches, particularly in education. Word problems, especially in subjects like mathematics, and the need to solve these problems by collectively addressing specific stages of reasoning, have raised the question of whether LLMs can be successful in this area as well. In our study, we conducted analyses by asking mathematics questions especially related to word problems using ChatGPT, which is based on the latest language models like Generative Pretrained Transformer (GPT). Additionally, we compared the correct and incorrect answers by posing the same questions to LLMMathChain, a mathematics-specific LLM based on the latest language models like LangChain. It was observed that the answers obtained were more successful with ChatGPT (GPT 3.5), particularly in the field of mathematics. However, both language models were found to be below expectations, particularly in word problems, and suggestions for improvement were provided. © 2023 ACM.
  • Data Paper
    Database Covering the Previously Excluded Daily Life Activities
    (2023) Mihçin, Şenay; Şahin, Ahmet Mert; Yılmaz, Mehmet; Alpkaya, Alican Tuncay; Tuna, Merve; Can, Nuray Korkmaz; Şahin, Serap; Akdeniz, Sevinç; Tosun, Aliye
    In biomedical engineering, implants are designed according to the boundary conditions of gait data and tested against. However, due to diversity in cultural backgrounds, religious rituals might cause different ranges of motion and different loading patterns. Especially in the Eastern part of the world, diverse Activities of Daily Living (ADL) consist of salat, yoga rituals, and different style sitting postures. Although databases cover ADL for the Western population, a database covering these diverse activities of the Eastern world, specific to these populations is non-existent. To include previously excluded ADL is a key step in understanding the kinematics and kinetics of these activities. By means of developments in motion capture technologies, excluded ADL data are captured to obtain the coordinate values to calculate the range of motion and the joint reaction forces. This study focuses on data collection protocol and the creation of an online database of previously excluded ADL activities, targeting 200 healthy subjects via Qualisys and IMU motion capture systems, and force plates, from West and Middle East Asian populations. Anthropometrics are known to affect kinematics and kinetics which are also included in the collected data. The current version of the database covers 50 volunteers for 12 different activities, the database aims for 100- male and 100- female healthy volunteers as the final target including C3D and BVH file types. The tasks are defined and listed in a table to create a database to make a query based on age, gender, BMI, type of activity and motion capture system. The data is collected only from a healthy population to understand healthy motion patterns during these previously excluded ADLs. The collected data is to be used for designing implants to allow these sorts of activities to be performed without compromising the quality of life of patients performing these activities in the future.
  • Conference Object
    Citation - Scopus: 1
    A Novel Feature To Predict Buggy Changes in a Software System
    (Springer, 2022) Yılmaz, Rahime; Nalçakan, Yağız; Haktanır, Elif
    Researchers have successfully implemented machine learning classifiers to predict bugs in a change file for years. Change classification focuses on determining if a new software change is clean or buggy. In the literature, several bug prediction methods at change level have been proposed to improve software reliability. This paper proposes a model for classification-based bug prediction model. Four supervised machine learning classifiers (Support Vector Machine, Decision Tree, Random Forrest, and Naive Bayes) are applied to predict the bugs in software changes, and performance of these four classifiers are characterized. We considered a public dataset and downloaded the corresponding source code and its metrics. Thereafter, we produced new software metrics by analyzing source code at class level and unified these metrics with the existing set. We obtained new dataset to apply machine learning algorithms and compared the bug prediction accuracy of the newly defined metrics. Results showed that our merged dataset is practical for bug prediction based experiments. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
  • Conference Object
    Deep Convolutional Neural Networks for Viability Analysis Directly From Cell Holograms Captured Using Lensless Holographic Microscopy
    (The Chemical and Biological Microsystems Society (CBMS), 2019) Delikoyun, Kerem; Çine, Ersin; Anıl İnevi, Müge; Özçivici, Engin; Özuysal, Mustafa; Tekin, Hüseyin Cumhur
    Cell viability analysis is one of the most widely used protocols in the fields of biomedical sciences. Traditional methods are prone to human error and require high-cost and bulky instrumentations. Lensless digital inline holographic microscopy (LDIHM) offers low-cost and high resolution imaging. However, recorded holograms should be digitally reconstructed to obtain real images, which requires intense computational work. We introduce a deep transfer learning-based cell viability classification method that directly processes the hologram without reconstruction. This new model is only trained once and viability of each cell can be predicted from its hologram. © 2019 CBMS-0001.
  • Conference Object
    Citation - Scopus: 1
    Distributed Identity Based Private Key Generation for Scada Systems
    (Springer, 2013) Kılınç, Görkem; Nai Fovino, Igor
    The security of the ICT (Information Communications Technology) components of industrial systems is gaining great importance in the context of their criticality for society at large. There is an urgent need for the consideration of security in their design, and for the analysis of the related vulnerabilities and potential threats. The high exposure of industrial critical infrastructure to such threats is mainly due to the intrinsic weakness of the communication protocols used to control the process network. The peculiarities of the industrial protocols (low computational power, large geographical distribution, near to real-time constraints) make hard the effective use of traditional cryptographic schemes and in particular the implementation of a effective key management infrastructure supporting a cryptographic layer. In this paper we present the first working prototype of a distributed key generation infrastructure for SCADA systems based on the well known identity based crypto-paradigm. © 2013 Springer-Verlag.
  • Conference Object
    The Performance Results of Ecdsa Implementation on Different Coordinate Systems
    (Tafford Publishing, 2008) Atay, Serap
    Elliptic Curve Cryptography has a high computational cost due to arithmetic operations of point addition and point doubling. But the cost can be reduced if different coordinate systems utilized. This paper shows that the performance of an elliptic curve digital signature algorithm (ECDSA) can be significantly increased by using different coordinate systems.
  • Conference Object
    A Memory Management Model for Cryptographic Software Libraries
    (Tafford Publishing, 2008) Mersin, Ali; Beyazıt, Mutlu
    Cryptographic protocols are implemented on the abstraction of multiple precision number libraries in which the dominant design criterion mostly turns out to be the maximization of the system performance. In contrast, each protocol may have its own memory usage pattern. In general case, the memory allocation and release routines are frequently called during the runtime. For this reason, an improper memory management strategy may yield an inefficient implementation. In this paper, we propose a memory management technique which is constructed under the consideration of the context of high level cryptographic software running on multi-programmed environments. Also, we show the implementation results of our approach and discuss with respect to the common static and dynamic memory allocation strategies. © 2008 Atilla Elçi.
  • Book Part
    Citation - Scopus: 2
    A Survey on Security in Wireless Sensor Networks: Attacks and Defense Mechanisms
    (IGI Global, 2013) Tekbacak, Fatih; Dalkılıç, Mehmet Emin; Korkmaz, İlker; Dağdeviren, Orhan
    Wireless Sensor Network (WSN) is a promising technology that has attracted the interest of research in the last decade. Security is one of the fundamental issues in sensor networks since sensor nodes are very resource constrained. An attacker may modify, insert, and delete new hardware and software components to the system where a single node, a specific part of the sensing area, and the whole network may become inoperable. Thus, the design of early attack detection and defense mechanisms must be carefully considered. In this chapter, the authors survey attacks and their defense mechanisms in WSNs. Attacks are categorized according to the related protocol layer. They also investigate the open research issues and emerging technologies on security in WSNs.
  • Article
    Citation - WoS: 4
    Citation - Scopus: 5
    Query optimization: Mobile agents versus accuracy of the cost estimation
    (CRL Publishing, 2005) Özakar, Belgin; Morvan, F.; Hameurlaint, A.
    Since an increasing number of diverse sources of data and information become available through World Wide Web, the field of distributed heterogeneous query processing attracts attention of the researchers. One of the main concerns is to reduce the amount of communication and the volume of data transferred in terms of query optimization where it is a real challenge to have the statistics of the resources predictable and up-to-date. Autonomy, proactivity and mobility features of mobile agents seem promising under some conditions. In this paper we are interested in the study of the efficiency of the mobile agents in relation with the approach of the cost model used during the optimization process. We present an execution model based on mobile agents. Performance evaluation shows the efficiency intervals of the execution model according to the estimation errors and the current state of the system. The major contribution of this paper is to point out the effective use of an execution model based on mobile agents in relation with the approach of the cost model and with the query type. © 2005 CRL Publishing Ltd.