{"title":"A Study on Power and Energy Measurement of NVIDIA Jetson Embedded GPUs Using Built-in Sensor","authors":"Büşra Aslan, Ayse Yilmazer-Metin","doi":"10.1109/UBMK55850.2022.9919522","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919522","url":null,"abstract":"Artificial intelligence (AI) has been shifted to the embedded devices known as edge devices. Component-level power is very important for the design and optimization of applications on edge devices to estimate energy consumption. Thus, accurate power measurements are needed for battery-powered systems. However, it is not straightforward. Because the behavior of a GPU is rather complex and not well documented. In this work, we report challenges getting power measurements using the built-in power sensor for an NVIDIA Jetson GPU device. We provide a method for true power and energy measurements of the kernels running on NVIDIA Jetson family GPUs.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"35 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113931616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Towards Third Generation AI: Explainable and Interpretable AI","authors":"Nilgün Sengöz, Tuncay Yiğit","doi":"10.1109/UBMK55850.2022.9919510","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919510","url":null,"abstract":"Today, artificial intelligence-based systems, especially machine learning and deep neural network algorithms, make decisions that directly affect people's lives, from health to autonomous vehicles and even to the defense sector. And yet, algorithms like deep neural networks that are high in performance accuracy but low in explainability and trust bring problems in ethics and interpretability. Explainable Artificial Intelligence (XAJD) is a branch of artificial intelligence that produces highquality interpretability, end-user-oriented understanding and explainable tools, techniques and algorithms. In this study, general information about XAI is given.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115689764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Türkay Şahi̇n, Tuna Çakar, Tunahan Bozkan, Seyit Ertugrul, A. Sayar
{"title":"Modeling Consumer Creditworthiness via Psychometric Scale and Machine Learning","authors":"Türkay Şahi̇n, Tuna Çakar, Tunahan Bozkan, Seyit Ertugrul, A. Sayar","doi":"10.1109/UBMK55850.2022.9919596","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919596","url":null,"abstract":"Although the predictive power of economic metrics to detect the creditworthiness of the customers is high, there is a rising interest in the integration of cognitive, psychological, behavioral, alternative, and demographic data into credit risk systems and processing the data through modern methods. The primary motivation for the rising interest is increased customer classification accuracy. In this research, customer creditworthiness was modeled through data consisting of personality, money attitudes, impulsivity, self-esteem, self-control, and material values and processed through artificial intelligence. The obtained findings have been evaluated as a reference point for the following research.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"893 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123251240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Addressing Class Imbalance for Transformer Based Knee MRI Classification","authors":"Gökay Sezen, I. Oksuz","doi":"10.1109/UBMK55850.2022.9919578","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919578","url":null,"abstract":"For assessing knee injuries, Magnetic Resonance Image (MRI) examinations are commonly utilized. Developing an automatic interpretable detection mechanism is an essential task for automating the clinical diagnosis of knee MRI. The imbalanced dataset problem is generally an issue for learning models in which the distribution of classes in the dataset is asymmetrical. The MRI datasets are generally imbalanced in favor of categories with injuries because patients who have an MRI are more likely to suffer a knee injury. Hence, it can be a challenging task to train a machine learning algorithm that can automatically handle class imbalance. In this paper, we propose both a network architecture and a comparison of the handling imbalanced dataset techniques to detect the general abnormalities in knee MR images. A network architecture that consists of CNN and transformer-based layers is proposed. Six different configuration methods for imbalanced data training are developed and compared with evaluation metrics (ROCAUC score, specificity, sensitivity, accuracy). Augmentation of additional data to the under-represented class and use of focal loss yield better classification specificity and AUC.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129412002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kadir Eker, Ayşe Gül Eker, Dilek Mandal, M. K. Pehlivanoglu, N. Duru
{"title":"Network Traffic Classification with Machine Learning Approaches","authors":"Kadir Eker, Ayşe Gül Eker, Dilek Mandal, M. K. Pehlivanoglu, N. Duru","doi":"10.1109/UBMK55850.2022.9919497","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919497","url":null,"abstract":"Today, detection of malware is often done automatically with artificial intelligence supported systems. In this way, this task, which was previously done manually, can be performed more precisely and more quickly. Darknet traffic is an internet network with many different cybercrime threats. It is also fake traffic observed in empty address space. A globally valid set of Internet Protocol (IP) addresses that are not assigned to any host or device. In this study, a deep learning model and pervasive machine learning classification algorithms are compared using the CIC-Darknet2020 dataset to describe Darknet traffic. The ‘Tor’ and ‘Vpn’ classes in the dataset, ‘Darknet’, ‘Non-Tor’ and ‘Non-Vpn’ classes are also specified as harmless. Various adjustments have been made for the unbalanced dataset, which has been performed with multiple classification, various machine learning algorithms. Random Forest Algorithm, Decision Trees and Deep Learning Model were the algorithms with the most successful accuracy values with %97.3, %97.4 %98.2values respectively.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129940731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Impact of Video Conferencing Applications on Remote Workers","authors":"Çağatay Aktaş, Mehmet Göktürk","doi":"10.1109/UBMK55850.2022.9919501","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919501","url":null,"abstract":"With the effect of the pandemic that has affected the worldwide in recent years, the spread of remote working has made it popular to conduct collaborative works such as pair programming, debugging and bug finding in software engineering through video conferencing applications. Although this situation makes it difficult for employees to synchronize and adapt with each other from time to time, the advantages of working remotely should not be ignored. In this study, some studies with eye tracking devices in the literature will be mentioned and the results will be summarized in order to measure the interaction of software teams with each other while working remotely with video conferencing applications. Then, the efficiency between face-toface working with the employees of a company that has adopted the semi-remote working methodology and remote paired programming using video conferencing applications will be compared.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126812428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Oguzhan Kizilbey, Murat Arslan, Yusuf Bayrak, Anil Çetinkaya, Aydin Yügrük, Erkan Danaci
{"title":"A Novel Software for Automatic Calibration Factor Measurement of RF Power Sensors","authors":"Oguzhan Kizilbey, Murat Arslan, Yusuf Bayrak, Anil Çetinkaya, Aydin Yügrük, Erkan Danaci","doi":"10.1109/UBMK55850.2022.9919539","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919539","url":null,"abstract":"In this letter, a software for calibration factor (CF) measurement of radio frequency (RF) power sensors (PS) by using VNA-based direct comparison transfer method (VBDCTM) was developed on C# platform. Measurements were performed between 10 MHz and 18 GHz frequency range. When the calibration factors calculated with conventional CF measurement method and novel software were compared, a maximum difference of 4.56% was found in the 10 MHz – 18 GHz frequency band. Therefore, the automatic CF measurement software based on VBDCTM, which has less complexity, can be preferred to conventional method for measuring the calibration factor of power sensors.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130463424","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
E. Adali, Sirojiddinov Shuhrat Samariddinovich, Khamroyeva Shahlo Mirdjonovna
{"title":"Comparison of Uzbek-Turkish Derivational Suffixes","authors":"E. Adali, Sirojiddinov Shuhrat Samariddinovich, Khamroyeva Shahlo Mirdjonovna","doi":"10.1109/ubmk55850.2022.9919509","DOIUrl":"https://doi.org/10.1109/ubmk55850.2022.9919509","url":null,"abstract":"In cases where there are few translated texts for two languages, lexical-based parallel corpus are used for computer translation. While this method is easy for fusuonal languages that can take very little affixes, it can be said to be more difficult for agglutinative languages. In agglutinative languages, it is important to know the meaning that the construction affixes give to the word. In this paper, we present comparatively the morphology of the Uzbek and Turkish languages, which we have prepared especially to create a word-based parallel corpus of Uzbek-Turkish languages and to make computer translations, in terms of construction suffixes.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124882287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Development of an Information Model of The Portal of Scientific Knowledge by Means of Semantic Web Technology","authors":"Zh.B. Sadirmekova, A. Yerimbetova, A. Ibraimkulov","doi":"10.1109/UBMK55850.2022.9919463","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919463","url":null,"abstract":"The approach to the development of the portal of scientific knowledge model is described. The main purpose of the portal is to provide meaningful access to scientific and educational information resources of this field of knowledge and comprehensive information processing services. The portal of scientific knowledge will provide systematization and integration of scientific knowledge, data and information resources into a single information space, meaningful and effective access to them, as well as support for their use in solving various scientific and educational tasks. One of the approaches to the creation of such portals is the use of Semantic Web technologies, relevant standards and tools in the collection, processing, storage, search and dissemination of information. It should be noted that our approach naturally integrates the most important components of Semantic Web technology, in particular, the use of ontology to represent the semantics of information resources and support their intellectual analysis.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125091795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Disaggregated Near-RT RIC Control Plane with Unified 5G DB for NS, MEC and NWDAF Integration","authors":"Hale Donertasli, Madhukiran Medithe","doi":"10.1109/UBMK55850.2022.9919594","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919594","url":null,"abstract":"The use-case of RAN Controller with MEC and Network Slicing integration by using unified 5G data collection and analytics is one of the key enablers for 5G networks to function based on the low latency requirements. Connecting these components are challenging for the operators due to the lack of single source of authority and unified data. O-RAN architecture extends the SDN controller concept of decoupling the control-plane (CP) from the user-plane (UP) by enhancing the traditional RAN functions with AI via RAN Intelligent Controller (RIC). However, disaggregating only CP and DP is not enough for the integration of the key enablers, indeed there is a need for disaggregation of CP of Near-RT RIC itself as well so that MEC, NS and NWDAF can connect in form of xApps which are the first anchor point of the authority and data handshake for time sensitive applications to be landed. In this paper, we proposed an architectural perspective focusing on the Near-RT RIC platform with the disaggregated CP enhancements and end-to-end combined xApp-rApp solution for smooth NS, MEC and NWDAF integration.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122240033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}