2022 7th International Conference on Computer Science and Engineering (UBMK)最新文献

{"title":"A Use Case on Evaluation of Productivity in Software Development Industry","authors":"Ecem Aktı, Can Balcı, H. Kilinç","doi":"10.1109/UBMK55850.2022.9919507","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919507","url":null,"abstract":"In this study, the productivity processes in an organization that develops software and manages data centers are discussed and the effect of automating daily operations on productivity is examined. The results of the productivity project realized for operation management were shared and the importance of automation in digitalization have been revealed. As the usecase scenario, cloud and microservice based next generation communication platform is used. Thanks to the automation developed with open source products for the monitoring of the platform, the tasks for which an engineer spends 2 weeks per month are carried out automatically. Before the productivity project, 144 hours a week were worked for 22 different products, after which this time decreased to 3.5 hours.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"10 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":"129153180","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":"Hybrid CPU-GPU Acceleration of a Multithreaded Image Stitching Algorithm","authors":"Shewit W. Tesfay, Zeynep G. Demirdag, H. F. Ugurdag, H. Ateş","doi":"10.1109/UBMK55850.2022.9919473","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919473","url":null,"abstract":"Real-time image stitching is critical, especially in un-manned aerial vehicles, and its acceleration has received attention in recent years. This paper describes an image stitching acceleration scheme for heterogeneous (CPU+GPU) devices. Acceleration is attempted with both multithreading and multiprocessing. Most time-critical functions in the algorithm are offloaded on to the GPU. We crafted a 3-buffer ping-pong mechanism for synchro-nization and data transfer among threads/processes in order to maximize CPU utilization. We carried out our experiments on Nvidia Jetson AGX Xavier. Results show that more than 3x acceleration is achieved.","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":"130776173","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":"Steel Surface Defect Classification Via Deep Learning","authors":"Mustafa Mert Tunal, A. Yıldız, Tuna Çakar","doi":"10.1109/UBMK55850.2022.9919470","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919470","url":null,"abstract":"Deep learning and image processing methods have taken place in many parts of our lives, as well as in the quality control stages of production lines. The aim of this study is to train and use a deep learning model to improve quality management using limited data and computing power. To achieve that, deep learning for quality control models were trained by classifying six different steel surface defect images in the NEU-DET dataset. Xception, ResNetV2 152, VGG19 and InceptionV3 architectures were used to train the model. High accuracy was obtained with both Xception and ResNetV2 152.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"18 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":"132517349","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":"Gas Cylinder Detection Using Deep Learning Based YOLOv5 Object Detection Method","authors":"Abdulkadir Albayrak, M. S. Özerdem","doi":"10.1109/UBMK55850.2022.9919478","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919478","url":null,"abstract":"Detection and tracking of objects has critical importance in terms of speeding up the process and facilitating the work in many areas. Especially in the process of counting objects, which is difficult and time-consuming for experts. In this paper, a study was carried out to detect gas cylinders with different colors and shapes using the deep learning-based Yolov5 method. The process of counting cylinders in the stock area or in the filling facilities can be difficult for the specialist due to the different sizes, arrangement and large number of cylinders. Within the scope of the study, a data set containing different types of cylinders in gas filling facilities was created. When the obtained results are evaluated, it has been observed that the Yolov5 algorithm detects the gas cylinders with different color and shape properties with a high success rate of 96.16%. In addition to the detection success, it has been observed that the method is also successful in different objective detections such as precision, sensitivity and box intersection.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"5 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":"130998105","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":"Reverse Ant Colony Optimization for the Winner Determination Problem in Combinatorial Auctions","authors":"Serhat Uzunbayir","doi":"10.1109/UBMK55850.2022.9919488","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919488","url":null,"abstract":"An auction is an effective process of trading items among bidders and sellers. Combinatorial auctions are auctions in which bidders can place bids on a bundle of items rather than bidding on a single item. As a result, they lead to more efficient allocations compared to traditional auctions. Determining the winners whose bids maximize the auctioneer's profit is known as the winner determination problem. The problem is NP-complete since it is not possible to solve it in polynomial time as the inputs increase. In this paper, reverse ant colony optimization algorithm is proposed for the problem which focuses on maximization of the ants' route instead of minimization of the regular version. The experimental results are compared using different size data sets with a previously proposed genetic algorithm and a random search algorithm. The experiments indicate that, as the search space expands, the proposed algorithm finds better solutions than the others.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"186 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":"131695178","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":"A Finite State Transducer Based Morphological Analyzer for The Kazakh Language","authors":"Gulmira Tolegen, Alymzhan Toleu, R. Mussabayev","doi":"10.1109/UBMK55850.2022.9919445","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919445","url":null,"abstract":"This paper presents a finite state transducer based morphological analyzer for Kazakh language which is able to decompose complex Kazakh words into consecutive morphemes including lemma, part-of-speech, and morphological tags. Due to the agglutinative nature of the language, the analyzer can produce more than one analysis for each word depending on word's complexity. We conducted several experiments to evaluate the performance of the analyzer. It achieved 92% coverage on large Wikipedia and 96% coverage on the news data, and the accuracy of analyzer was 98.40% on the test data.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"2 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":"116218578","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":"Classifying Turkish Trade Registry Gazette Announcements","authors":"İrem Nur Demirtaş, Seçcil Arslan, Gülşen Eryiğit","doi":"10.1109/UBMK55850.2022.9919536","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919536","url":null,"abstract":"Turkish Trade Registry Gazette is an important source of information in many sectors such as banking and telecommunication. Although the newspaper is publicly available, the data is hard to acquire, and announcements are offered in image format. It is possible to search for a specific announcement a company has, but there exist many other unrelated announcements in the image returned. This poses multiple challenges in the way of information extraction. Due to the structure of the documents in these images, it is hard to perform OCR directly. Moreover, even in the case where the text is extracted, the announcement boundaries must be detected to split the announcements within the page. Once the announcements are extracted, the announcement of the searched company should be matched. Since no information regarding the surrounding announcements is given as a result of the query, these announcements should also be categorized to detect any events of interest other companies may have. In this work, we address all of these problems and present a pipeline that includes image processing, OCR, announcement splitting, and document classification steps.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"55 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":"121361018","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":"Darknet Traffic Classification with Machine Learning Algorithms and SMOTE Method","authors":"Hasan Karagöl, Oğuzhan Erdem, Barkin Akbas, Tuncay Soylu","doi":"10.1109/UBMK55850.2022.9919462","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919462","url":null,"abstract":"The Darknet is a network that can be accessed with certain privileges and runs a non-standard communication protocol. The Darknet traffic that consists of data from several known networks such as Tor and the P2P is often used for criminal activities due to its anonymity. It is so critical to correctly classify Darknet traffic to differentiate the individual flows for security purposes. In this paper, we proposed three different machine learning (ML) based traffic classification approaches; the binary classification of Darknet and Benign traffic classes (Case 1); the quadruple classification of classes Tor, NonTor, VPN, and NonVpn (Case 2); an traffic classification of eight sub-traffic classes (Case 3). We further applied the SMOTE method for balancing the sizes of the classes in the traffic dataset and feature selection (FS) algorithms to identify the most effective attributes where the number of features in the original dataset were reduced from 63 to 8, 8 and 6 for Case 1, 2 and 3 respectively. For all three cases, classification was performed with six different machine learning algorithms with and without SMOTE, and the highest accuracy values were obtained with SMOTE method. The highest accuracy values were obtained with the Random Forest Algorithm as 97.22%, 97.16% and 85.99% for Case 1, 2 and 3, respectively.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"53 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":"122493238","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":"Comparison of Heuristic and Metaheuristic Algorithms","authors":"A. Tunc, Şakir Taşdemir, T. Sağ","doi":"10.1109/UBMK55850.2022.9919459","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919459","url":null,"abstract":"The development of computer technologies day by day has enabled the use of new technological approaches in solving many problems. Technological innovations are used to solve many problems, especially thanks to hardware devices that develop in capacity, and algorithms that continue to develop rapidly. These technologies are used to provide significant time and/or performance gains in solving problems. The great change in software technologies has led to the emergence of many smart algorithms. With the development of artificial intelligence approaches, algorithms inspired by many nature and natural events have been used in the solution stages of problems. These approaches, which take the intuitive movements of living things as an example, have revealed a new solution approach in addition to the mathematical and statistical models used for problem-solving. These algorithms, which are called heuristic algorithms, aim to create the most appropriate solution set by considering the time and/or performance gains of the solution sets. The use of these algorithms, which can be used in solving many problems from production to design, from optimization problems to classification problems, is quite common. With the development of heuristic algorithms, new approaches such as meta-heuristic and hyper-heuristic algorithms have been introduced. In our study, a detailed examination has been made of these algorithms, which are classified as heuristic algorithms, and especially heuristic algorithms have been compared with meta-heuristic (meta-heuristic) algorithms, and details on their similarities and differences have been tried to be presented. Algorithms are shown by classifying them according to their structures. In particular, the basic features of heuristic and meta-heuristic algorithms such as search space, performance, workspace, search behaviors, search process, simplicity, reliability, flexibility, and initial requirements have been examined and the similarities and differences between them have been tried to be shown with examples. Information on current algorithms published in recent years and their applicability for solving problems are also given.","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":"132712828","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":"Forecasting the Short-Term Electricity In Steel Manufacturing For Purchase Accuracy on Day-Ahead Market","authors":"A. Koca, Z. Erdem, Mehmet N. Aydin","doi":"10.1109/UBMK55850.2022.9919563","DOIUrl":"https://doi.org/10.1109/UBMK55850.2022.9919563","url":null,"abstract":"Forecasting electricity consumption in the most accurate way is crucial for purchase on the day-ahead market in steel manufacturing. This study is aimed to predict short-term electricity consumption regarding the day-ahead market purchase by employing important features of electricity consumption time-series data. We utilize Random Forest (RF), Gradient-Boosted Trees (GBT), and Generalized Linear Models (GLM), as they are appropriate for the given problem and widely used regression algorithms for prediction purposes. This study leverages the regression algorithms in the Apache Spark Machine Learning library. The performance of the prediction models is evaluated based on the standard deviation of the residuals (RMSE) and the proportion of variance explained (R-squared). We additionally discuss the distribution of prediction errors of the models. Experiments show that the RF model outperforms the GBT and GLM. It is considered that the results can contribute to accurate forecasting of short-term electricity demand for purchasing on the day-ahead.","PeriodicalId":417604,"journal":{"name":"2022 7th International Conference on Computer Science and Engineering (UBMK)","volume":"51 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113985828","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}