Yazar "Bilgin, Turgay Tugay" seçeneğine göre listele
Listeleniyor 1 - 9 / 9
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A Classification Approach with Machine Learning Methods for Technical Problems of Distance Education: Turkey Example(INT COUNCIL OPEN & DISTANCE EDUCATION, 2021) Yayla, Rıdvan; Yayla, Halime Nur; Ortaç, Gizem; Bilgin, Turgay TugayDistance education is an education model in which the lessons can be taught simultaneously using technical material without time and space restrictions. It has gained importance after the Covid-19 pandemic processes and has been implemented as a valid educational model in all educational institutions. Due to the sudden pandemic measures, distance education has brought about a lot of technical problems at unprepared educational institutions against the pandemic. In this paper, a classification approach is proposed by machine learning methods on Twitter instead of the usual structured research methods such as survey, one-on-one meeting for technical problems of distance education. The most encountered and commented distance education problem, which can be defined in different languages by the proposed method, have been analysed with Turkey example. Sentiment analysis has been made from negative and neutral tweets about distance education. The problems have been classified by natural language processing methods based on Turkish word analysis.Öğe Classification of Turkish Universities by quantity and titles of Academic Staff(Ieee, 2018) Kaşif, Ahmet; Bilgin, Turgay TugayIn this study, universities are clustered based on the quantity and title of the academics employed by Turkish Universities. The data comes from both state and private universities and vocational schools. For clustering, partitional and hierarchical clustering methods were used. The dataset was obtained from the YOK academic search website using "web scraping" techniques implemented by the authors. Agglomerative clustering technique was found to yield better results regarding cluster sizes and intracluster distances. The Silhouette coefficient is used for clustering quality measurement. As a result of the study, Turkish universities were mainly composed of four clusters. These are instructor dominant universities, just instructor based vocational schools, professor and associate professor dominated universities and research assistant dominated universities. The results of this study may be used for academic staff employment planning in Turkish Universities.Öğe Determining of the user attitudes on mobile security programs with machine learning methods(Slovene Society Informatika, 2021) Yayla, Rıdvan; Bilgin, Turgay TugaySecurity plays an important role in today's virtual world. Cybersecurity software has been widely used by the development of portable virtual environments. Smartphones occur in an important part of our lives. Daily routines are performed over mobile phones, especially after the COVID-19 pandemic process. Due to its ease of use, compulsory or optional mobile phone use also brought about many security concerns. Mobile security software is used for different purposes such as virus removal and protection of personal information according to user preferences. In the field of natural language processing, user preferences can now be analyzed on the basis of machine learning methods with sentiment analysis. In this paper, the preference reasons for mobile security software have been analysed with machine learning methods based on user comments and sentiment analysis. In the study, all user comments have been classified into 10 main categories and the user preferences of mobile security programs have been analysed.Öğe Improving Initial Flattening of Convex-Shaped Free-Form Mesh Surface Patches Using a Dynamic Virtual Boundary(C R L Publishing Ltd, 2019) Yavuz, Erdem; Yazici, Rifat; Kasapbasi, Mustafa Cem; Bilgin, Turgay TugayThis study proposes an efficient algorithm for improving flattening result of triangular mesh surface patches having a convex shape. The proposed approach, based on barycentric mapping technique, incorporates a dynamic virtual boundary, which considerably improves initial mapping result. The dynamic virtual boundary approach is utilized to reduce the distortions for the triangles near the boundary caused by the nature of convex combination technique. Mapping results of the proposed algorithm and the base technique are compared by area and shape accuracy metrics measured for several sample surfaces. The results prove the success of the proposed approach with respect to the base method.Öğe The Investigation of the Success of Different Machine Learning Methods in Breast Cancer Diagnosis(Düzce Üniversitesi, 2021) Ates, Ibrahim; Bilgin, Turgay TugayObjective: The aim of this study is to identify cancer earlier in life using machine learning methods. Methods: For this purpose, the Wisconsin Diagnostic Breast Cancer dataset was classified using Naive Bayes, decision trees, artificial neural networks algorithms and comparison of these machine learning methods was made. KNIME Analytics Platform was used for applications. Before the classification process, the dataset was preprocessed. After the preprocessing stage, three different classifier methods were applied to the dataset. Accuracy, sensitivity, specificity and confusion matrices were used to measure the success of the methods. Results: The results show that Naive Bayes and artificial neural network methods classify tumors with 96.5% accuracy. The success of the decision tree method in classification was 92.6%. Conclusions: The machine learning algorithms can be used successfully in breast cancer diagnosis to determine whether the tumors are malign or benign.Öğe A New Approach to Minimize Memory Requirements of Frequent Subgraph Mining Algorithms(Gazi Universitesi, 2021) Bilgin, Turgay TugayFrequent subgraph mining (FSM) is a subsection of graph mining domain which is extensively used for graph classification and graph clustering purposes. Over the past decade, many efficient FSM algorithms have been developed. The improvements generally focus on reducing time complexity by changing the algorithm structure or using parallel programming techniques. FSM algorithms have another problem to solve, which is the high memory consumption. In this study, a new approach called Predictive Dynamic Sized Structure Packing (PDSSP) have been proposed to minimize the memory requirement of FSM algorithms. Proposed approach redesigns the internal data structures of FSM algorithms without any algorithmic modifications. PDSSP has two contributions. The first one is the Dynamic Sized Integer Type (ds_Int) which is a newly designed unsigned integer data type. The second contribution is "Data Structure packaging" component that uses a data structure packing technique which changes the behaviour of the compiler. A number of experiments have been conducted to examine the effectiveness and efficiency of the PDSSP approach by embedding it into two state-of-art algorithms called gSpan and Gaston. Proposed implementation have been compared to the official one. Almost all results show that the proposed implementation consumes less memory on each support level. As a result, PDSSP extensions can save memory and the peak memory usage may decrease up to 38% depending on the dataset.Öğe Novel Approach to Minimize the Memory Requirements of Frequent Subgraph Mining Techniques(John Wiley and Sons Inc, 2021) Bilgin, Turgay Tugay; Murat, OğuzFrequent subgraph mining (FSM) is a subset of the graph mining domain that is extensively used for graph classification and clustering. Over the past decade, many efficient FSM algorithms have been developed with improvements generally focused on reducing the time complexity by changing the algorithm structure or using parallel programming techniques. FSM algorithms also require high memory consumption, which is another problem that should be solved. In this paper, we propose a new approach called Predictive dynamic sized structure packing (PDSSP) to minimize the memory needs of FSM algorithms. Our approach redesigns the internal data structures of FSM algorithms without making algorithmic modifications. PDSSP offers two contributions. The first is the Dynamic Sized Integer Type, a newly designed unsigned integer data type, and the second is a data structure packing technique to change the behavior of the compiler. We examined the effectiveness and efficiency of the PDSSP approach by experimentally embedding it into two state-of-the-art algorithms, gSpan and Gaston. We compared our implementations to the performance of the originals. Nearly all results show that our proposed implementation consumes less memory at each support level, suggesting that PDSSP extensions could save memory, with peak memory usage decreasing up to 38% depending on the dataset.Öğe Performing similarity analysis on organic farming crop data of Turkish Cities(Institute of Electrical and Electronics Engineers Inc., 2020) Kaşif, Ahmet; Ortaç, Gizem; Ibis, E.; Bilgin, Turgay TugayOrganic crop production is an important technique which can help increase the quality and throughput of food production. In this study, similarity analysis of organic farming crop data of Turkish cities has been performed. The dataset has been collected from the web site of Turkish Ministry of Agriculture and Forestry. Some pre-processing techniques have been applied. Dynamic Time Warping distance has been used as a similarity metric. Results show that Dynamic Time Warping similarity is suitable for similarity detection of organic crop production. © 2020 IEEE.Öğe Web Proxy Log Data Mining System for Clustering Users and Search Keywords(2017) Bilgin, Turgay Tugay; Aytekin, Mustafa KorayIn this study, Internet users were clustered by the search keywords which they type into search bars of search engines. Our proposed software is called UQCS (User Queries Clustering System) and it is developed to demonstrate the efficiency of our hypothesis. UQCS co-operates with the Strehl’s relationship based clustering toolkit and performs segmentation on users based on the keywords they use for searching the web. Internet Proxy server logs were parsed and query strings were extracted from the search engine URL’s and the resulting IP-Term matrix was converted into a similarity matrix using Euclidean, Jaccard, Cosine Distance and Pearson Correlation Distance metrics. K- Means and graph-based OPOSSUM algorithm were used to perform clustering on the similarity matrices. Results were illustrated by using CLUSION visualization toolkit.
