Feriştah Dalkılıç | Dokuz Eylül University (original) (raw)

Papers by Feriştah Dalkılıç

Lecture notes on data engineering and communications technologies, 2020

International Journal of Multidisciplinary Studies and Innovative Technologies, Dec 31, 2022

Data visualization, which is essentially the visual expression of data through mathematical calcu... more Data visualization, which is essentially the visual expression of data through mathematical calculations, is a highly effective method for transferring and understanding data. The right data visualization, selected or developed with data and people in mind, has an important role in improving the quality of conveying the relationships, meanings, information, structures and hidden properties of data to individuals from different fields of expertise and enables them to work together. Data visualization has been a subject that has evolved and progressed in different ways throughout history and has finally reached today's computer technology and conditions. The improved facilities of today's computer such as high computing power has brought data visualization methods to a new horizon. Also the changing definition of the data has expanded the boundaries of the data visualization. Today, data is obtained from many different sources such as; IoT, Embedded systems, Social data, Business data and Real-Life data. This study is also inspired by the Real-Life, and a visualization system was developed based on real-life botanical trees and using humans as data. Throughout history, computer science and other branches of science have used art as a subject in various studies. This study reunites science and art and proposes a way of expressing the human through art. Thus, an artistic visualization system was developed that uses the activities of people's lives as data and generates artistic visuals inspired by abstract art paintings. Through this study, a sub-art style was created with a visualization system that manages to produce artistic visuals by adhering to mathematical foundations.

2022 Innovations in Intelligent Systems and Applications Conference (ASYU), Sep 7, 2022

Scientific Programming, Aug 8, 2018

e dimensionality reduction and visualization problems associated with multivariate centroids obta... more e dimensionality reduction and visualization problems associated with multivariate centroids obtained by clustering algorithms are addressed in this paper. Two approaches are used in the literature for the solution of such problems, specifically, the self-organizing map (SOM) approach and mapping selected two features manually (MS2Fs). In addition, principle component analysis (PCA) was evaluated as a component for solving this problem on supervised datasets. Each of these traditional approaches has drawbacks: if SOM runs with a small map size, all centroids are located contiguously rather than at their original distances according to the high-dimensional structure; MS2Fs is not an efficient method because it does not take features outside of the method into account, and lastly, PCA is a supervised method and loses the most valuable feature. In this study, five novel hybrid approaches were proposed to eliminate these drawbacks by using the quantum genetic algorithm (QGA) method and four feature selection methods, Pearson's correlation, gain ratio, information gain, and relief methods. Experimental results demonstrate that, for 14 datasets of different sizes, the prediction accuracy of the proposed weighted clustering approaches is higher than the traditional K-means++ clustering approach. Furthermore, the proposed approach combined with K-means++ and QGA shows the most efficient placements of the centroids on a two-dimensional map for all the test datasets.

nternational journal of advanced research in computer and communication engineering, Jul 30, 2016

While the Internet is evolving to the Internet of Things, all other technologies that are related... more While the Internet is evolving to the Internet of Things, all other technologies that are related to it are also advancing in a way to contain and support concepts like device-to-device networking, proximate discovery, energy efficiency and security. One of the fastest thrives can be seen in mobile communication technologies. This became more obvious after 4G spread out. In our study, we present a review of a new and revolutionary mobile technology under development: LTE Direct; which runs on licensed radio spectrum and is claimed to be energy efficient and secure, while enabling new approaches for the Internet of Things. We state why and how LTE Direct should replace existing systems, by making an analysis considering provided features, resource consumption, possible use cases and security concerns, as well as comparisons with the conventional technologies. Lastly, we provide ideas for the areas where further research should be made to have this system be a reality.

Journal of Advanced Transportation, 2017

Planning a journey by integrating route and timetable information from diverse sources of transpo... more Planning a journey by integrating route and timetable information from diverse sources of transportation agencies such as bus, ferry, and train can be complicated. A user-friendly, informative journey planning system may simplify a plan by providing assistance in making better use of public transportation. In this study, we presented the service-oriented, multimodel Intelligent Journey Planning System, which we developed to assist travelers in journey planning. We selected Izmir, Turkey, as the pilot city for this system. The multicriteria problem is one of the well-known problems in transportation networks. Our study proposes a gradual path-finding algorithm to solve this problem by considering transfer count and travel time. The algorithm utilizes the techniques of efficient algorithms including round based public transit optimized router, transit node routing, and contraction hierarchies on transportation graph. We employed Dijkstra's algorithm after the first stage of the path-finding algorithm by applying stage specific rules to reduce search space and runtime. The experimental results show that our path-finding algorithm takes 0.63 seconds of processing time on average, which is acceptable for the user experience.

Twitter is a social network, which contains information of the city events (concerts, festival, e... more Twitter is a social network, which contains information of the city events (concerts, festival, etc.), city problems (traffic, collision, and road incident), the news, feelings of people, etc. For these reasons, there are many studies, which use tweet data to detect useful information to support the smart city management. In this paper, the ways of finding citizen problems with their locations by using tweet data is discussed. Tweets in Turkish language from the Aegean Region of Turkey were used for the study. It is aimed to form a smart system, which detects problems of citizens and extracts the problems' exact locations from tweet texts. Firstly, the collected data was analyzed to get information of any city event, citizen's complaint or requests about a problem. After the possibility of detecting tweets, which have any city problem, was ensured, two datasets were created. The first one consists of the tweets that have an event information or a problem and the second one has the tweets, which have other information not related to our study. Then Naive Bayes classifier was trained on the annotated tweets and was tested on a separate set of tweets. Accuracy, precision, recall, and F-measure of the classifier is given. A location recognizer, which finds the Turkish place names in a text, is created and applied on the tweets that are marked as information-containing by the classifier to detect the location of the problem precisely. The first findings of the project is promising. The high accuracy, which is obtained by the classifier, shows that it is proper to use this classifier for our study. The location recognizer is planned to be improved and place names on the real-time tweet data is to be detected.

Applied sciences, Aug 6, 2022

This article is an open access article distributed under the terms and conditions of the Creative... more This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY

All protein targets of a compound might not be identified during the compound development stage. ... more All protein targets of a compound might not be identified during the compound development stage. The expected side effects of compounds while using them in treatments might be observed due to the binding of compounds to off-target proteins and the biological processes triggered by these off-targets. If the protein targets of compounds would be identified more comprehensively, the side effects observed after a disease treatment might be also reduced. The aim of this study is to identify potential targets of a compound with a computational method. The proposed method wm compute potential off-targets of a compound by using gene expression data of compound-treated cells on protein-protein interaction networks. This method mimics the cellular processes in terms of topological relations by integrating protein interactions and the transcriptome data of a given compound. The method first maps simplified compound effects on the network, and then computes various network centrality metrics to suggest the most probable targets of a new compound. The experiments revealed that the type of interaction network dramatically effects the target identification performance of the method. Furthermore, network centrality metrics might produce variable results based on selected confidence cut-offs and the network type. The proposed method simply implements graph-based data mining on integrated biological resources to reduce time and cost of computer-based compound development.

International Journal of Multidisciplinary Studies and Innovative Technologies, 2022

Author identification is one of the application areas of text mining. It deals with the automatic... more Author identification is one of the application areas of text mining. It deals with the automatic prediction of the potential author of an electronic text among predefined author candidates by using author specific writing styles. In this study, we conducted an experiment for the identification of the author of a Turkish language text by using classical machine learning methods including Support Vector Machines (SVM), Gaussian Naive Bayes (GaussianNB), Multi Layer Perceptron (MLP), Logistic Regression (LR), Stochastic Gradient Descent (SGD) and ensemble learning methods including Extremely Randomized Trees (ExtraTrees), and eXtreme Gradient Boosting (XGBoost). The proposed method was applied on three different sizes of author groups including 10, 15 and 20 authors obtained from a new dataset of newspaper articles. Term frequency-inverse document frequency (TF-IDF) vectors were created by using 1-gram and 2-gram word tokens. Our results show that the most successful method is the SGD with a classification performance accuracy of 0.976% by using word unigrams and most successful method is the LR with a classification performance accuracy of 0.935% by using word bigrams.

Yükseköğretim ve Bilim Dergisi, 2017

Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrenc... more Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrencilerin devamsızlık nedenlerinin ortaya çıkarılmasını amaçlayan ve devamsızlığın öğrencilerin başarı durumları üzerindeki etkisinin incelendiği bu araştırmada, veri madenciliği tekniklerinden biri olan Apriori algoritması kullanılarak birliktelik kuralları çıkartılmıştır. Araştırma sonucunda, cinsiyetin, devam edilmekte bölümün ve akademik yılın, örgün ya da ikinci öğretimde okuyor olmanın, genel başarı durumunun, devamsızlık eğilimi üzerinde etkileri olduğu saptanmıştır. Ayrıca, fakülteden memnuniyet, ikamet edilen yer ve bu yerin fakülteye olan uzaklığı devamsızlık davranışlarını etkilerken, fakülteye ulaşım şeklinin devamsızlık davranışı üzerine önemli bir etkisinin olmadığı gözlenmiştir.

DergiPark (Istanbul University), Dec 31, 2021

Today, due to the intense use of social media platforms such as Twitter by all segments of today'... more Today, due to the intense use of social media platforms such as Twitter by all segments of today's technology, people have begun to share their views, ideas, and feelings through these media. It is possible to discover mighty valuable knowledge from this enormous resource. This study has emerged to assist users in making choices by evaluating emotions about TV series and movies that have recently appeared on social platforms, using ideas and feelings. The textual tweet data was preprocessed and cleaned of noise by using natural language processing techniques. Tweets were tagged using the Bert-based model according to the content of the Turkish TV series and movie comments, and their polarities were calculated. Machine learning models including Naïve Bayes (NB), Support Vector Machines (SVM), Random Forest (RF); Bagging and Voting, which are among the general ensemble algorithms, were trained for sentiment analysis by taking the obtained polarity values. The voting algorithm gives the best accuracy at 87%, while the Support Vector Machines give the best area under the receiver operating characteristics curve (AUC) of 0.96. A web application was developed by using Flask to monitor sentiment scores via hashtags (#).

Journal of Medical Imaging and Health Informatics, May 1, 2018

Lecture notes on data engineering and communications technologies, 2020

Factorization of large integers has been being considered as a challenging problem in computer sc... more Factorization of large integers has been being considered as a challenging problem in computer science and engineering since the earliest times of the computer technology. Despite the comprehensive efforts, there is still no reported deterministic polynomial-time algorithm; however, its complexity class is in fact not yet decided. A fast and robust polynomial-time algorithm for this problem is required to increase the processing capabilities of current systems. Yet, there are also hesitations at the same time within the community, due to the potential security threats that may appear in such a case. The (asymptotically) fastest algorithm ever found so far to factor large integers is the general number field sieve. Its performance depends on selection of “good” polynomials, which requires a specific procedure for such a selection. Another significant performance factor surely is the power of the processing hardware and their peripherals. This article unveils and discusses the impacts of heterogeneous computing using a graphics processor units (GPU) instead of a central processing unit (CPU) on the performance of polynomial selection and so of factoring large integers. Further, the article presents implementation details and a comparative performance evaluation of the Base-m polynomial selection method to select “good” polynomials. Accordingly, the GPU is found to be more effective over larger numbers with more roots, while the CPU appeared more effective over smaller numbers with less roots, possibly due to the excessive overheads in the GPU processing procedures.

Celal Bayar Universitesi Fen Bilimleri Dergisi, Mar 22, 2019

Global positioning system and other outdoor positioning mechanisms are already subject to compreh... more Global positioning system and other outdoor positioning mechanisms are already subject to comprehensive research and development for almost half a century. Conversely, indoor positioning services became a hot topic in the last decade. Since GPS (and. other outdoor solutions) do not work reliably in most indoor environments, researchers and developers are working on accurate positioning solutions, especially tailored for indoor places. However; due to walls, furniture, people and other obstacles, absolute location estimation is very hard and expensive to achieve in indoor places. In addition, accuracy needs depend on the scenario and application. In this study, we have studied the feasibility of room-level location detection in home and office environments. We have focused on examining the quality of room-wise detection accuracy of the fingerprinting method that is applied along with standard Wi-Fi radio infrastructure. We have conducted experiments in a multi-storey office building made of concrete and aerated concrete bricks with many rooms, in which it is significantly hard to accurately estimate the correct place of a thing, using radio signals. To the best of our knowledge, our paper is the first study that investigates the room-level accuracy of Wi-Fi fingerprinting-based indoor localization systems. We have found out that, it is possible to feasibly achieve room-level detection with good accuracy, via a pre-calculated room-specific received signal strength indicator threshold value.

Yükseköðretim ve bilim dergisi, 2017

Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrenc... more Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrencilerin devamsızlık nedenlerinin ortaya çıkarılmasını amaçlayan ve devamsızlığın öğrencilerin başarı durumları üzerindeki etkisinin incelendiği bu araştırmada, veri madenciliği tekniklerinden biri olan Apriori algoritması kullanılarak birliktelik kuralları çıkartılmıştır. Araştırma sonucunda, cinsiyetin, devam edilmekte bölümün ve akademik yılın, örgün ya da ikinci öğretimde okuyor olmanın, genel başarı durumunun, devamsızlık eğilimi üzerinde etkileri olduğu saptanmıştır. Ayrıca, fakülteden memnuniyet, ikamet edilen yer ve bu yerin fakülteye olan uzaklığı devamsızlık davranışlarını etkilerken, fakülteye ulaşım şeklinin devamsızlık davranışı üzerine önemli bir etkisinin olmadığı gözlenmiştir.

Bugun, dunyadaki su kaynaklarinin tukenmesiyle birlikte temiz su ihtiyacinin artacagi ongorulmekt... more Bugun, dunyadaki su kaynaklarinin tukenmesiyle birlikte temiz su ihtiyacinin artacagi ongorulmektedir. Tarim alaninda bilincsiz sulama hizla temiz su kaynaklari tuketmektedir. Ayrica tarim urunlerindeki verimi azaltmaktadir. Kuresel isinmanin etkileri ile su daha degerli hale gelmistir. Teknolojinin gelismesiyle birlikte Nesnelerin Interneti (IoT) tum alanlarda yayilmaya baslamistir. Ustun karar verme, bilgisayarlarin insanlardan daha gelismis olmasi ve gelisime acik olmasiyla saglanabilir. Topraktaki nem, sicaklik ve mineral degerleri cok kucuk toleranslarla olculebilir ve buna gore cikarimlar yapilabilir. Insanlarin bitkilere ihtiyac duyulan miktarda suyu vermeleri cok zor olmasina ragmen, bu islem bilgisayarlar tarafindan yapilabilir. Bu calismada, Nesnelerin interneti teknolojisini yapay zeka ile birlestirmek icin calismalar yapilmistir. Mikrodenetleyici ve duyargalarin yardimiyla elde edilen bilgiler makine ogrenimi ile islenmis ve gelecekteki durumlar icin otomatik karar verme yapisi olusturulmustur. Bu yazida elde edilen kazanclar; Sulama alaninda bilincsiz su kullanimindan kacinmak, sulamayi en uygun sekilde yaparak uretim verimliligini arttirmak, su kaybini ve dolayisiyla uretim alanindaki maliyetleri azaltmak, kullanilan insan gucunu en aza indirmek ve insan zayifliklarindan kaynaklanan hatalari onlemek, sulama yapilacak alana fiziksel olarak ulasmanin zor oldugu yerlerde uzaktan erisim ile kullanici kontrolunu saglamaktir.

arXiv (Cornell University), Jul 12, 2019

It is not hard to see that the need for clean water is growing by considering the decrease of the... more It is not hard to see that the need for clean water is growing by considering the decrease of the water sources day by day in the world. Potable fresh water is also used for irrigation, so it should be planned to decrease fresh water wastage. With the development of the technology and the availability of cheaper and more effective solutions, the efficiency of the irrigation increased and the water loss can be reduced. In particular, Internet of things (IoT) devices have begun to be used in all areas. We can easily and precisely collect temperature, humidity and mineral values from the irrigation field with the IoT devices and sensors. Most of the operations and decisions about irrigation are carried out by people. For people, it is hard to have all the real time data such as temperature, moisture and mineral levels in the decision-making process and make decisions by considering them. People usually make decisions with their experience. In this study, a wide range of information from irrigation field was obtained by using IoT devices and sensors. Data collected from IoT devices and sensors sent via communication channels and stored on MongoDB. With the help of Weka software, the data was normalized and the normalized data was used as a learning set. As a result of the examinations, decision tree (J48) algorithm with the highest accuracy was chosen and artificial intelligence model was created. Decisions are used to manage operations such as starting, maintaining and stopping the irrigation. The accuracy of the decisions was evaluated and the irrigation system was tested with the results. There are options to manage, view the system remotely and manually and also see the system's decisions with the created mobile application.

Scientific Programming, 2018

The dimensionality reduction and visualization problems associated with multivariate centroids ob... more The dimensionality reduction and visualization problems associated with multivariate centroids obtained by clustering algorithms are addressed in this paper. Two approaches are used in the literature for the solution of such problems, specifically, the self-organizing map (SOM) approach and mapping selected two features manually (MS2Fs). In addition, principle component analysis (PCA) was evaluated as a component for solving this problem on supervised datasets. Each of these traditional approaches has drawbacks: if SOM runs with a small map size, all centroids are located contiguously rather than at their original distances according to the high-dimensional structure; MS2Fs is not an efficient method because it does not take features outside of the method into account, and lastly, PCA is a supervised method and loses the most valuable feature. In this study, five novel hybrid approaches were proposed to eliminate these drawbacks by using the quantum genetic algorithm (QGA) method an...

Yükseköğretim ve Bilim Dergisi, 2017

Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrenc... more Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrencilerin devamsızlık nedenlerinin ortaya çıkarılmasını amaçlayan ve devamsızlığın öğrencilerin başarı durumları üzerindeki etkisinin incelendiği bu araştırmada, veri madenciliği tekniklerinden biri olan Apriori algoritması kullanılarak birliktelik kuralları çıkartılmıştır. Araştırma sonucunda, cinsiyetin, devam edilmekte bölümün ve akademik yılın, örgün ya da ikinci öğretimde okuyor olmanın, genel başarı durumunun, devamsızlık eğilimi üzerinde etkileri olduğu saptanmıştır. Ayrıca, fakülteden memnuniyet, ikamet edilen yer ve bu yerin fakülteye olan uzaklığı devamsızlık davranışlarını etkilerken, fakülteye ulaşım şeklinin devamsızlık davranışı üzerine önemli bir etkisinin olmadığı gözlenmiştir.

Lecture notes on data engineering and communications technologies, 2020

International Journal of Multidisciplinary Studies and Innovative Technologies, Dec 31, 2022

Data visualization, which is essentially the visual expression of data through mathematical calcu... more Data visualization, which is essentially the visual expression of data through mathematical calculations, is a highly effective method for transferring and understanding data. The right data visualization, selected or developed with data and people in mind, has an important role in improving the quality of conveying the relationships, meanings, information, structures and hidden properties of data to individuals from different fields of expertise and enables them to work together. Data visualization has been a subject that has evolved and progressed in different ways throughout history and has finally reached today's computer technology and conditions. The improved facilities of today's computer such as high computing power has brought data visualization methods to a new horizon. Also the changing definition of the data has expanded the boundaries of the data visualization. Today, data is obtained from many different sources such as; IoT, Embedded systems, Social data, Business data and Real-Life data. This study is also inspired by the Real-Life, and a visualization system was developed based on real-life botanical trees and using humans as data. Throughout history, computer science and other branches of science have used art as a subject in various studies. This study reunites science and art and proposes a way of expressing the human through art. Thus, an artistic visualization system was developed that uses the activities of people's lives as data and generates artistic visuals inspired by abstract art paintings. Through this study, a sub-art style was created with a visualization system that manages to produce artistic visuals by adhering to mathematical foundations.

2022 Innovations in Intelligent Systems and Applications Conference (ASYU), Sep 7, 2022

Scientific Programming, Aug 8, 2018

e dimensionality reduction and visualization problems associated with multivariate centroids obta... more e dimensionality reduction and visualization problems associated with multivariate centroids obtained by clustering algorithms are addressed in this paper. Two approaches are used in the literature for the solution of such problems, specifically, the self-organizing map (SOM) approach and mapping selected two features manually (MS2Fs). In addition, principle component analysis (PCA) was evaluated as a component for solving this problem on supervised datasets. Each of these traditional approaches has drawbacks: if SOM runs with a small map size, all centroids are located contiguously rather than at their original distances according to the high-dimensional structure; MS2Fs is not an efficient method because it does not take features outside of the method into account, and lastly, PCA is a supervised method and loses the most valuable feature. In this study, five novel hybrid approaches were proposed to eliminate these drawbacks by using the quantum genetic algorithm (QGA) method and four feature selection methods, Pearson's correlation, gain ratio, information gain, and relief methods. Experimental results demonstrate that, for 14 datasets of different sizes, the prediction accuracy of the proposed weighted clustering approaches is higher than the traditional K-means++ clustering approach. Furthermore, the proposed approach combined with K-means++ and QGA shows the most efficient placements of the centroids on a two-dimensional map for all the test datasets.

nternational journal of advanced research in computer and communication engineering, Jul 30, 2016

While the Internet is evolving to the Internet of Things, all other technologies that are related... more While the Internet is evolving to the Internet of Things, all other technologies that are related to it are also advancing in a way to contain and support concepts like device-to-device networking, proximate discovery, energy efficiency and security. One of the fastest thrives can be seen in mobile communication technologies. This became more obvious after 4G spread out. In our study, we present a review of a new and revolutionary mobile technology under development: LTE Direct; which runs on licensed radio spectrum and is claimed to be energy efficient and secure, while enabling new approaches for the Internet of Things. We state why and how LTE Direct should replace existing systems, by making an analysis considering provided features, resource consumption, possible use cases and security concerns, as well as comparisons with the conventional technologies. Lastly, we provide ideas for the areas where further research should be made to have this system be a reality.

Journal of Advanced Transportation, 2017

Planning a journey by integrating route and timetable information from diverse sources of transpo... more Planning a journey by integrating route and timetable information from diverse sources of transportation agencies such as bus, ferry, and train can be complicated. A user-friendly, informative journey planning system may simplify a plan by providing assistance in making better use of public transportation. In this study, we presented the service-oriented, multimodel Intelligent Journey Planning System, which we developed to assist travelers in journey planning. We selected Izmir, Turkey, as the pilot city for this system. The multicriteria problem is one of the well-known problems in transportation networks. Our study proposes a gradual path-finding algorithm to solve this problem by considering transfer count and travel time. The algorithm utilizes the techniques of efficient algorithms including round based public transit optimized router, transit node routing, and contraction hierarchies on transportation graph. We employed Dijkstra's algorithm after the first stage of the path-finding algorithm by applying stage specific rules to reduce search space and runtime. The experimental results show that our path-finding algorithm takes 0.63 seconds of processing time on average, which is acceptable for the user experience.

Twitter is a social network, which contains information of the city events (concerts, festival, e... more Twitter is a social network, which contains information of the city events (concerts, festival, etc.), city problems (traffic, collision, and road incident), the news, feelings of people, etc. For these reasons, there are many studies, which use tweet data to detect useful information to support the smart city management. In this paper, the ways of finding citizen problems with their locations by using tweet data is discussed. Tweets in Turkish language from the Aegean Region of Turkey were used for the study. It is aimed to form a smart system, which detects problems of citizens and extracts the problems' exact locations from tweet texts. Firstly, the collected data was analyzed to get information of any city event, citizen's complaint or requests about a problem. After the possibility of detecting tweets, which have any city problem, was ensured, two datasets were created. The first one consists of the tweets that have an event information or a problem and the second one has the tweets, which have other information not related to our study. Then Naive Bayes classifier was trained on the annotated tweets and was tested on a separate set of tweets. Accuracy, precision, recall, and F-measure of the classifier is given. A location recognizer, which finds the Turkish place names in a text, is created and applied on the tweets that are marked as information-containing by the classifier to detect the location of the problem precisely. The first findings of the project is promising. The high accuracy, which is obtained by the classifier, shows that it is proper to use this classifier for our study. The location recognizer is planned to be improved and place names on the real-time tweet data is to be detected.

Applied sciences, Aug 6, 2022

This article is an open access article distributed under the terms and conditions of the Creative... more This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY

All protein targets of a compound might not be identified during the compound development stage. ... more All protein targets of a compound might not be identified during the compound development stage. The expected side effects of compounds while using them in treatments might be observed due to the binding of compounds to off-target proteins and the biological processes triggered by these off-targets. If the protein targets of compounds would be identified more comprehensively, the side effects observed after a disease treatment might be also reduced. The aim of this study is to identify potential targets of a compound with a computational method. The proposed method wm compute potential off-targets of a compound by using gene expression data of compound-treated cells on protein-protein interaction networks. This method mimics the cellular processes in terms of topological relations by integrating protein interactions and the transcriptome data of a given compound. The method first maps simplified compound effects on the network, and then computes various network centrality metrics to suggest the most probable targets of a new compound. The experiments revealed that the type of interaction network dramatically effects the target identification performance of the method. Furthermore, network centrality metrics might produce variable results based on selected confidence cut-offs and the network type. The proposed method simply implements graph-based data mining on integrated biological resources to reduce time and cost of computer-based compound development.

International Journal of Multidisciplinary Studies and Innovative Technologies, 2022

Author identification is one of the application areas of text mining. It deals with the automatic... more Author identification is one of the application areas of text mining. It deals with the automatic prediction of the potential author of an electronic text among predefined author candidates by using author specific writing styles. In this study, we conducted an experiment for the identification of the author of a Turkish language text by using classical machine learning methods including Support Vector Machines (SVM), Gaussian Naive Bayes (GaussianNB), Multi Layer Perceptron (MLP), Logistic Regression (LR), Stochastic Gradient Descent (SGD) and ensemble learning methods including Extremely Randomized Trees (ExtraTrees), and eXtreme Gradient Boosting (XGBoost). The proposed method was applied on three different sizes of author groups including 10, 15 and 20 authors obtained from a new dataset of newspaper articles. Term frequency-inverse document frequency (TF-IDF) vectors were created by using 1-gram and 2-gram word tokens. Our results show that the most successful method is the SGD with a classification performance accuracy of 0.976% by using word unigrams and most successful method is the LR with a classification performance accuracy of 0.935% by using word bigrams.

Yükseköğretim ve Bilim Dergisi, 2017

Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrenc... more Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrencilerin devamsızlık nedenlerinin ortaya çıkarılmasını amaçlayan ve devamsızlığın öğrencilerin başarı durumları üzerindeki etkisinin incelendiği bu araştırmada, veri madenciliği tekniklerinden biri olan Apriori algoritması kullanılarak birliktelik kuralları çıkartılmıştır. Araştırma sonucunda, cinsiyetin, devam edilmekte bölümün ve akademik yılın, örgün ya da ikinci öğretimde okuyor olmanın, genel başarı durumunun, devamsızlık eğilimi üzerinde etkileri olduğu saptanmıştır. Ayrıca, fakülteden memnuniyet, ikamet edilen yer ve bu yerin fakülteye olan uzaklığı devamsızlık davranışlarını etkilerken, fakülteye ulaşım şeklinin devamsızlık davranışı üzerine önemli bir etkisinin olmadığı gözlenmiştir.

DergiPark (Istanbul University), Dec 31, 2021

Today, due to the intense use of social media platforms such as Twitter by all segments of today'... more Today, due to the intense use of social media platforms such as Twitter by all segments of today's technology, people have begun to share their views, ideas, and feelings through these media. It is possible to discover mighty valuable knowledge from this enormous resource. This study has emerged to assist users in making choices by evaluating emotions about TV series and movies that have recently appeared on social platforms, using ideas and feelings. The textual tweet data was preprocessed and cleaned of noise by using natural language processing techniques. Tweets were tagged using the Bert-based model according to the content of the Turkish TV series and movie comments, and their polarities were calculated. Machine learning models including Naïve Bayes (NB), Support Vector Machines (SVM), Random Forest (RF); Bagging and Voting, which are among the general ensemble algorithms, were trained for sentiment analysis by taking the obtained polarity values. The voting algorithm gives the best accuracy at 87%, while the Support Vector Machines give the best area under the receiver operating characteristics curve (AUC) of 0.96. A web application was developed by using Flask to monitor sentiment scores via hashtags (#).

Journal of Medical Imaging and Health Informatics, May 1, 2018

Lecture notes on data engineering and communications technologies, 2020

Factorization of large integers has been being considered as a challenging problem in computer sc... more Factorization of large integers has been being considered as a challenging problem in computer science and engineering since the earliest times of the computer technology. Despite the comprehensive efforts, there is still no reported deterministic polynomial-time algorithm; however, its complexity class is in fact not yet decided. A fast and robust polynomial-time algorithm for this problem is required to increase the processing capabilities of current systems. Yet, there are also hesitations at the same time within the community, due to the potential security threats that may appear in such a case. The (asymptotically) fastest algorithm ever found so far to factor large integers is the general number field sieve. Its performance depends on selection of “good” polynomials, which requires a specific procedure for such a selection. Another significant performance factor surely is the power of the processing hardware and their peripherals. This article unveils and discusses the impacts of heterogeneous computing using a graphics processor units (GPU) instead of a central processing unit (CPU) on the performance of polynomial selection and so of factoring large integers. Further, the article presents implementation details and a comparative performance evaluation of the Base-m polynomial selection method to select “good” polynomials. Accordingly, the GPU is found to be more effective over larger numbers with more roots, while the CPU appeared more effective over smaller numbers with less roots, possibly due to the excessive overheads in the GPU processing procedures.

Celal Bayar Universitesi Fen Bilimleri Dergisi, Mar 22, 2019

Global positioning system and other outdoor positioning mechanisms are already subject to compreh... more Global positioning system and other outdoor positioning mechanisms are already subject to comprehensive research and development for almost half a century. Conversely, indoor positioning services became a hot topic in the last decade. Since GPS (and. other outdoor solutions) do not work reliably in most indoor environments, researchers and developers are working on accurate positioning solutions, especially tailored for indoor places. However; due to walls, furniture, people and other obstacles, absolute location estimation is very hard and expensive to achieve in indoor places. In addition, accuracy needs depend on the scenario and application. In this study, we have studied the feasibility of room-level location detection in home and office environments. We have focused on examining the quality of room-wise detection accuracy of the fingerprinting method that is applied along with standard Wi-Fi radio infrastructure. We have conducted experiments in a multi-storey office building made of concrete and aerated concrete bricks with many rooms, in which it is significantly hard to accurately estimate the correct place of a thing, using radio signals. To the best of our knowledge, our paper is the first study that investigates the room-level accuracy of Wi-Fi fingerprinting-based indoor localization systems. We have found out that, it is possible to feasibly achieve room-level detection with good accuracy, via a pre-calculated room-specific received signal strength indicator threshold value.

Yükseköðretim ve bilim dergisi, 2017

Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrenc... more Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrencilerin devamsızlık nedenlerinin ortaya çıkarılmasını amaçlayan ve devamsızlığın öğrencilerin başarı durumları üzerindeki etkisinin incelendiği bu araştırmada, veri madenciliği tekniklerinden biri olan Apriori algoritması kullanılarak birliktelik kuralları çıkartılmıştır. Araştırma sonucunda, cinsiyetin, devam edilmekte bölümün ve akademik yılın, örgün ya da ikinci öğretimde okuyor olmanın, genel başarı durumunun, devamsızlık eğilimi üzerinde etkileri olduğu saptanmıştır. Ayrıca, fakülteden memnuniyet, ikamet edilen yer ve bu yerin fakülteye olan uzaklığı devamsızlık davranışlarını etkilerken, fakülteye ulaşım şeklinin devamsızlık davranışı üzerine önemli bir etkisinin olmadığı gözlenmiştir.

Bugun, dunyadaki su kaynaklarinin tukenmesiyle birlikte temiz su ihtiyacinin artacagi ongorulmekt... more Bugun, dunyadaki su kaynaklarinin tukenmesiyle birlikte temiz su ihtiyacinin artacagi ongorulmektedir. Tarim alaninda bilincsiz sulama hizla temiz su kaynaklari tuketmektedir. Ayrica tarim urunlerindeki verimi azaltmaktadir. Kuresel isinmanin etkileri ile su daha degerli hale gelmistir. Teknolojinin gelismesiyle birlikte Nesnelerin Interneti (IoT) tum alanlarda yayilmaya baslamistir. Ustun karar verme, bilgisayarlarin insanlardan daha gelismis olmasi ve gelisime acik olmasiyla saglanabilir. Topraktaki nem, sicaklik ve mineral degerleri cok kucuk toleranslarla olculebilir ve buna gore cikarimlar yapilabilir. Insanlarin bitkilere ihtiyac duyulan miktarda suyu vermeleri cok zor olmasina ragmen, bu islem bilgisayarlar tarafindan yapilabilir. Bu calismada, Nesnelerin interneti teknolojisini yapay zeka ile birlestirmek icin calismalar yapilmistir. Mikrodenetleyici ve duyargalarin yardimiyla elde edilen bilgiler makine ogrenimi ile islenmis ve gelecekteki durumlar icin otomatik karar verme yapisi olusturulmustur. Bu yazida elde edilen kazanclar; Sulama alaninda bilincsiz su kullanimindan kacinmak, sulamayi en uygun sekilde yaparak uretim verimliligini arttirmak, su kaybini ve dolayisiyla uretim alanindaki maliyetleri azaltmak, kullanilan insan gucunu en aza indirmek ve insan zayifliklarindan kaynaklanan hatalari onlemek, sulama yapilacak alana fiziksel olarak ulasmanin zor oldugu yerlerde uzaktan erisim ile kullanici kontrolunu saglamaktir.

arXiv (Cornell University), Jul 12, 2019

It is not hard to see that the need for clean water is growing by considering the decrease of the... more It is not hard to see that the need for clean water is growing by considering the decrease of the water sources day by day in the world. Potable fresh water is also used for irrigation, so it should be planned to decrease fresh water wastage. With the development of the technology and the availability of cheaper and more effective solutions, the efficiency of the irrigation increased and the water loss can be reduced. In particular, Internet of things (IoT) devices have begun to be used in all areas. We can easily and precisely collect temperature, humidity and mineral values from the irrigation field with the IoT devices and sensors. Most of the operations and decisions about irrigation are carried out by people. For people, it is hard to have all the real time data such as temperature, moisture and mineral levels in the decision-making process and make decisions by considering them. People usually make decisions with their experience. In this study, a wide range of information from irrigation field was obtained by using IoT devices and sensors. Data collected from IoT devices and sensors sent via communication channels and stored on MongoDB. With the help of Weka software, the data was normalized and the normalized data was used as a learning set. As a result of the examinations, decision tree (J48) algorithm with the highest accuracy was chosen and artificial intelligence model was created. Decisions are used to manage operations such as starting, maintaining and stopping the irrigation. The accuracy of the decisions was evaluated and the irrigation system was tested with the results. There are options to manage, view the system remotely and manually and also see the system's decisions with the created mobile application.

Scientific Programming, 2018

The dimensionality reduction and visualization problems associated with multivariate centroids ob... more The dimensionality reduction and visualization problems associated with multivariate centroids obtained by clustering algorithms are addressed in this paper. Two approaches are used in the literature for the solution of such problems, specifically, the self-organizing map (SOM) approach and mapping selected two features manually (MS2Fs). In addition, principle component analysis (PCA) was evaluated as a component for solving this problem on supervised datasets. Each of these traditional approaches has drawbacks: if SOM runs with a small map size, all centroids are located contiguously rather than at their original distances according to the high-dimensional structure; MS2Fs is not an efficient method because it does not take features outside of the method into account, and lastly, PCA is a supervised method and loses the most valuable feature. In this study, five novel hybrid approaches were proposed to eliminate these drawbacks by using the quantum genetic algorithm (QGA) method an...

Yükseköğretim ve Bilim Dergisi, 2017

Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrenc... more Öz Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi'nde öğrenim görmekte olan öğrencilerin devamsızlık nedenlerinin ortaya çıkarılmasını amaçlayan ve devamsızlığın öğrencilerin başarı durumları üzerindeki etkisinin incelendiği bu araştırmada, veri madenciliği tekniklerinden biri olan Apriori algoritması kullanılarak birliktelik kuralları çıkartılmıştır. Araştırma sonucunda, cinsiyetin, devam edilmekte bölümün ve akademik yılın, örgün ya da ikinci öğretimde okuyor olmanın, genel başarı durumunun, devamsızlık eğilimi üzerinde etkileri olduğu saptanmıştır. Ayrıca, fakülteden memnuniyet, ikamet edilen yer ve bu yerin fakülteye olan uzaklığı devamsızlık davranışlarını etkilerken, fakülteye ulaşım şeklinin devamsızlık davranışı üzerine önemli bir etkisinin olmadığı gözlenmiştir.