DSpace Arşivi :: by Yazar "Ozel, Selma Ayse" değerine göre listeleniyor

Yazar "Ozel, Selma Ayse" seçeneğine göre listele

Listeleniyor 1 - 20 / 20

A novel distance-based moving average model for improvement in the predictive accuracy of financial time series
(Elsevier, 2024) Ejder, Ugur; Ozel, Selma Ayse
Time-series forecasting is essential for system analysis. Many traditional studies have paid attention to individual stock-oriented solutions and disregarded general approaches on financial time series or skipped the dynamics of the system and its triggering components. It is difficult to fully adapt to evolving market conditions with stable financial indicators. For this reason, the proposed novel distance-based exponential moving-average (DBEMA) model is dynamically designed to overcome the changing conditions of financial time series. A novel distancebased moving-average feature model can produce an adaptive prediction approach for financial time series. To evaluate the impact of the novel proposed DBEMA features, they are compared to the features selected by recursive feature elimination using classification and regression trees among the financial indicators, using benchmark classification models. To confirm the performance of the proposed novel distance-based movingaverage features, the forecasting results of the features are compared using linear regression, bagged trees regressor, Gaussian naive Bayes, k-nearest neighbors, random forests, multilayer perceptron, convolutional neural network, long short-term memory, gated recurrent unit, and relative strength index method benchmark models. The experimental analysis has shown that methods with our proposed novel DBEMA features has better forecasting accuracy with respect to the methods without DBEMA. Therefore, the proposed novel distance-based moving-average methodology designed for financial time-series analysis demonstrates that it guides a new perspective in nonlinear time-series trends.
A Review of Distance Learning and Learning Management Systems
(Intech Europe, 2016) Keles, Mumine Kaya; Ozel, Selma Ayse
In recent years, rapid developments in technology and the web have led to many changes in education. One of the most important changes in education is in the form of distance learning. Distance learning, which is used to define education where educators and learners are physically separated, is not a new concept; however, emerging technologies and the web allow web-based distance learning and therefore increase its popularity. As a result of these developments, many universities have started to use web-based distance learning systems to provide flexible education that is independent of time and place. In this chapter, we review all popular, widely used, and well-known learning management systems and include detailed comparison of some of these systems to allow institutions to choose the right system for their distance education activities.
Comparison of Feature Selection Methods for Sentiment Analysis on Turkish Twitter Data
(IEEE, 2017) Parlar, Tuba; Sarac, Esra; Ozel, Selma Ayse
The Internet and social media provide a major source of information about people's opinions. Due to the rapidly growing number of online documents, it becomes both time-consuming and hard task to obtain and analyze the desired opinionated information. Sentiment analysis is the classification of sentiments expressed in documents. To improve classification perfromance feature selection methods which help to identify the most valuable features are generally applied. In this paper, we compare the performance of four feature selection methods namely Chi-square, Information Gain, Query Expansion Ranking, and Ant Colony Optimization using Maximum Entropi Modeling classification algorithm over Turkish Twitter dataset. Therefore, the effects of feature selection methods over the performance of sentiment analysis of Turkish Twitter data are evaluated. Experimental results show that Query Expansion Ranking and Ant Colony Optimization methods outperform other traditional feature selection methods for sentiment analysis.
Comparison of feature selection methods for sentiment analysis on Turkish Twitter data
(Institute of Electrical and Electronics Engineers Inc., 2017) Parlar, Tuba; Sarac, Esra; Ozel, Selma Ayse
The Internet and social media provide a major source of information about people's opinions. Due to the rapidly growing number of online documents, it becomes both time-consuming and hard task to obtain and analyze the desired opinionated information. Sentiment analysis is the classification of sentiments expressed in documents. To improve classification perfromance feature selection methods which help to identify the most valuable features are generally applied. In this paper, we compare the performance of four feature selection methods namely Chi-square, Information Gain, Query Expansion Ranking, and Ant Colony Optimization using Maximum Entropi Modeling classification algorithm over Turkish Twitter dataset. Therefore, the effects of feature selection methods over the performance of sentiment analysis of Turkish Twitter data are evaluated. Experimental results show that Query Expansion Ranking and Ant Colony Optimization methods outperform other traditional feature selection methods for sentiment analysis. © 2017 IEEE.
Deep Learning-based Sentiment Analysis of Facebook Data: The Case of Turkish Users
(Oxford Univ Press, 2021) Coban, Onder; Ozel, Selma Ayse; Inan, Ali
Sentiment analysis (SA) is an essential task for many domains where it is crucial to know users' public opinion about events, products, brands, politicians and so on. Existing works on SA have concentrated on English texts including Twitter feeds and user reviews on hotels, movies and products. On the other hand, Facebook, as an online social network (OSN), has attracted quite limited attention from the research community. Among these, SA work on Turkish text obtained from OSNs are extremely scarce. In this paper, our aim is to perform SA on public Facebook data collected from Turkish user accounts. Our study differs from existing studies in terms of the data set scale, the natural language of the texts in the data set and the extent of experimental analyses that include both machine learning and deep learning techniques. We extensively report not only the results of different learning models involving SA but also statistical distribution of metadata of user activities across various user attributes (e.g. gender and age). Our experimental results indicate that recurrent neural networks achieve the best accuracy (i.e. 0.916) with word embeddings. To the best of our knowledge, this is the best result for SA on Facebook data in the context of the Turkish language.
Detection and Cross-domain Evaluation of Cyberbullying in Facebook Activity Contents for Turkish
(Assoc Computing Machinery, 2023) Coban, Onder; Ozel, Selma Ayse; Inan, Ali
Cyberbullying refers to bullying and harassment of defenseless or vulnerable people such as children, teenagers, and women through any means of communication (e.g., e-mail, text messages, wall posts, tweets) over any online medium (e.g., social media, blogs, online games, virtual reality environments). The effect of cyberbullying may be severe and irreversible and it has become one of the major problems of cyber-societies in today's electronic world. Prevention of cyberbullying activities as well as the development of timely response mechanisms require automated and accurate detection of cyberbullying acts. This study focuses on the problem of cyberbullying detection over Facebook activity content written in Turkish. Through extensive experiments with the various machine and deep learning algorithms, the best estimator for the task is chosen and then employed for both cross-domain evaluation and profiling of cyber-aggressive users. The results obtained with fivefold cross-validation are evaluated with an average-macro F1 score. These results show that BERT is the best estimator with an average macro F1 of 0.928, and employing it on various datasets collected from different OSN domains produces highly satisfying results. This article also reports detailed profiling of cyber-aggressive users by providing even more information than what is visible to the naked eye.
Detection of Cyberbullying on Social Media Messages in Turkish
(IEEE, 2017) Ozel, Selma Ayse; Sarac, Esra; Akdemir, Seyran; Aksu, Hulya
The increased use of the Internet and the ease of access to online communities like social media have provided an avenue for cybercrimes. Cyberbullying, which is a kind of cybercrime, is defined as an aggressive, intentional action against a defenseless person by using the Internet, social media, or other electronic contents. Researchers have found that many of the bullying cases have tragically ended in suicides; hence automatic detection of cyberbullying has become important. The aim of this study is to detect cyberbullying on social media messages written in Turkish. To our knowledge, this is the first study which makes cyberbully detection on Turkish texts. We prepare a dataset from Instagram and Twitter messages written in Turkish and then we applied machine learning techniques that are Support Vector Machines (SVM), decision tree (C4.5), Naive Bayes Multinomial, and k Nearest Neighbors (kNN) classifiers to detect cyberbullying. We also apply information gain and chi-square feature selection methods to improve the accuracy of classifiers. We observe that when both words and emoticons in the text messages are taken into account as features, cyberbully detection improves. Among the classifiers, Naive Bayes Multinomial is the most successful one in terms both classification accuracy and running time. When feature selection is applied classification accuracy improves up to 84% for the dataset used.
Facebook Tells Me Your Gender: An Exploratory Study of Gender Prediction for Turkish Facebook Users
(Assoc Computing Machinery, 2021) Coban, Onder; Inan, Ali; Ozel, Selma Ayse
Online Social Networks (OSNs) are very popular platforms for social interaction. Data posted publicly over OSNs pose various threats against the individual privacy of OSN users. Adversaries can try to predict private attribute values, such as gender, as well as links/connections. Quantifying an adversary's capacity in inferring the gender of an OSN user is an important first step towards privacy protection. Numerous studies have been made on the problem of predicting the gender of an author/user, especially in the context of the English language. Conversely, studies in this field are quite limited for the Turkish language and specifically in the domain of OSNs. Previous studies for gender prediction of Turkish OSN users have mostly been performed by using the content of tweets and Facebook comments. In this article, we propose using various features, not just user comments, for the gender prediction problem over the Facebook OSN. Unlike existing studies, we exploited features extracted from profile, wall content, and network structure, as well as wall interactions of the user. Therefore, our study differs from the existing work in the broadness of the features considered, machine learning and deep learning methods applied, and the size of the OSN dataset used in the experimental evaluation. Our results indicate that basic profile information provides better results; moreover, using this information together with wall interactions improves prediction quality. We measured the best accuracy value as 0.982, which was obtained by combining profile data and wall interactions of Turkish OSN users. In the wall interactions model, we introduced 34 different features that provide better results than the existing content-based studies for Turkish.
Fine-grained Kinship Detection for Facebook Users based on Wall Contents
(Institute of Electrical and Electronics Engineers Inc., 2021) Ooban, Onder; Inan, Ali; Ozel, Selma Ayse
This paper investigates whether it is possible to automatically detect fine-grained kinship (not to detect its existence but to detect the type of kinship like child, father, grandfather, and so on) between two Facebook users or not. To do so, we present and employ a lexicon-based approach that completely depends on the wall contents of users. To the best of our knowledge, this is the first study towards kinship detection for both the type of input data (i.e., free OSN text) and the language (i.e., Turkish). We perform our experiments on a crawled snapshot of public Facebook data collected from accounts of users in Turkey. Our results are promising and show that a content-based approach can be a good starting point for future works even though it has some challenges. © 2021 IEEE.
Integrating an Online Compiler and a Plagiarism Detection Tool into the Moodle Distance Education System for Easy Assessment of Programming Assignments
(Wiley, 2015) Kaya, Mumine; Ozel, Selma Ayse
In this study, an online compiler and a source code plagiarism detection tool have been included into the Moodle based distance education system of our Computer Engineering department. For this purpose Moodle system has been extended with the GCC compiler, and the Moss source code plagiarism detection tool. We observed that using the online compiler and the plagiarism detection tool reduces time and effort needed for the assessment of the programming assignments; prevents our students from plagiarism; and increases their success in their programming based Data Structures course. (c) 2014 Wiley Periodicals, Inc. Comput Appl Eng Educ 23:363-373, 2015; View this article online at ; DOI
Inverse document frequency-based sensitivity scoring for privacy analysis
(Springer London Ltd, 2022) Coban, Onder; Inan, Ali; Ozel, Selma Ayse
Privacy risk analysis of online social network (OSN) users aims at generating a risk score for each OSN user such that higher scores potentially imply a greater risk of privacy violation. Privacy risk analysis is typically carried out over a response matrix (R) where any matrix element r(ij) indicates the portion of the OSN that the user i shares his/her attribute j. Most of the existing work relies on the mathematical framework of item response theory to derive sensitivity and visibility components from R. In this study, we propose interpreting R to be a term-document matrix and consequently suggest using the inverse document frequency (IDF) method as the sensitivity component. Experiments performed on both synthetic and real-world datasets show that the proposed IDF-based method can be used as a sensitivity component.
Multi-Classification of Depression Levels Based on Blood Biomarkers
(Institute of Electrical and Electronics Engineers Inc., 2024) Kavak, Rahmi; Ozel, Selma Ayse; Polat, Sema; Ozler, Sinan
Depression is a psychiatric condition characterized by a persistent feeling of sadness and diminished interest in significant activities. Diagnosing this health problem is challenging because it relies on several social and physiological factors. The timely identification of depression aids in the prevention of severe outcomes, such as suicide. Early detection of depression levels is crucial to prevent adverse effects and enhance the quality of everyday life. The purpose of this work is to use data preprocessing techniques to create a clean dataset from noisy and incomplete medical data that includes blood biomarker values of patients and then apply data mining techniques to this dataset to predict the degree of depression. Adana Dr. Ekrem Tok Mental Health Hospital supplied the raw data for the study, with consent from the ethics committee. Missing data are completed by filling with the constant value and the most frequent value methods (i.e., 0 and mode values) in the relevant column. The classification of the dataset is performed using AdaBoost, Decision Tree (DT), and Logistic Regression (LR) classifiers, which have been previously used in medical datasets and demonstrated to be effective. The Logistic Regression classifier achieved the highest success rate (Accuracy: 0.541 and weighted F-score: 0.481). © 2024 IEEE.
Named Entity Recognition over FBNER: A New Facebook Dataset in Turkish
(Institute of Electrical and Electronics Engineers Inc., 2021) Ooban, Onder; Ozel, Selma Ayse; Inan, Ali
In this paper, we introduce a new Named Entity Recognition (NER) dataset of Facebook messages written in the Turkish language. We also employ a Conditional Random Fields based NER system to discover named entities from Facebook messages. Our system achieves an F1 score of 0.713 when training and test sets include Facebook posts. We also obtained an F1 score of 0.599 when the training set is from the news domain. A strength of this research is that it is one of the first studies in this field that focuses on NER over Turkish Facebook messages. This is because performing NER on user-generated content turns into a very challenging task since such informal contents are often noisy texts that have arammatical and spelling errors. © 2021 IEEE.
Prediction of Students' Academic Success Using Data Mining Methods
(IEEE, 2018) Uzel, Vahide Nida; Turgut, Sultan Sevgi; Ozel, Selma Ayse
Success is very important for all of us. Most people wants prosperity, reputation, and richness that can only be achieved with the success. A society that wants to be successful should pay attention to their new generation because they are the future of the world. If we want to invest to our future, we must contribute to success of our new generations. Therefore, in this study, the academic performance of the students that belong to different levels of education like primary, secondary, and high school levels is tried to be determined by applying various classification methods such as Multilayer Perceptron (MLP), Random Forest (RF), Naive Bayes (NB), Decision Tree (J48), and Voting classifiers. It is also observed which characteristics are more related to the improvement of academic performance of the students. Features like absence of student, parent's school satisfaction, raising hands on class, and parent who is responsible for the student can affect the success of the student. A comparison is made with other study that previously worked on the same data set. As a result, better classification accuracy is achieved. We observe the best classification accuracy as 80.6% by Voting classifier, while the previous study has the highest accuracy as 79.1% by applying Artificial Neural Network (ANN) classifier. Also, in our study, Apriori algorithm is applied to detect relationships between features.
Privacy Risk Analysis for Facebook Users
(IEEE, 2020) Coban, Onder; Inan, Ali; Ozel, Selma Ayse
Disclosing personal information over Online Social Networks (OSNs) poses security and privacy risks. The privacy risk imposed on an OSN account can be measured through the privacy preferences and the network position of the corresponding account. In this study, two state-of-the-art methods are used to measure privacy risk of Turkish users of the Facebook OSN. Experimental results show that male users and users at age range 21-40 are at greater risk.
Similarity Detection between Turkish Text Documents with Distance Metrics
(IEEE, 2017) Kaya Keles, Mumine; Ozel, Selma Ayse
The aim of this study is to compare the successes of various distance metrics and to determine the most appropriate methods in order to detect similarities among textual documents written in Turkish. Computing similarities between text documents is the basic step of plagiarism detection, and text mining methods like author detection, text classification and clustering. Therefore, plagiarism detection and text mining applications will be more successful by using the distance metrics that are determined according to the results obtained in this study. For this purpose, chunks of texts in different lengths are selected as the experimental dataset in this study. After that, preprocessing methods are applied to the dataset that is used; therefore new and different experimental scenarios are created by removing stopwords and Turkish characters, and stemming words with Zemberek. According to the experimental results, it is observed that the preprocessing phase increases the accuracy of similarity detection. Especially, stemming using Zemberek increases the success rate. In all cases, the Cosine Similarity method has been observed as more successful than other distance metrics, because of producing more realistic results.
Transfer Learning with Fuzzy for Breast Cancer
(Inst Information Science, 2024) Hepsag, Pinar Uskaner; Ozel, Selma Ayse; Yazici, Adnan
Deep learning methods have been used to reduce the number of unnecessary breast biopsies. In this study, an accurate hybrid rule-based fuzzy system with transfer learning is developed to classify breast abnormalities as malignant or benign by calculating breast cancer risk from digital mammogram images. Our system consists of three phases: (i) data augmentation methods (e.g., traditional methods, Generative Adversarial Networks (GANs)); (ii) classification of the breast abnormalities on the BCDR-D02 and mini-MIAS databases by fine-tuning transfer learning methods with the deep learning base model Convolutional Neural Network (CNN); and (iii) calculation of breast cancer risk with a rule-based fuzzy system using the results of the second phase to improve the classification of breast abnormality results. Using our CNN baseline model and traditional extension methods, we achieve 64% and 82% accuracy for mini-MIAS and BCDR-D02, respectively. With fine-tuning the transfer learning methods, we obtain 80% and 83% with VGG-16 for mini- MIAS and BCDR-D02, respectively. Using the rule-based fuzzy system, called the risk method, we achieve the highest results for mini-MIAS (93%) and BCDR-D02 (94%). The classification results of our risk method are compared with the other transfer learning and baseline methods, and it is found that the accuracy of breast abnormality classification is improved by using a hybrid rule-based fuzzy system with transfer learning. Our study can serve as a guide that provides useful tips to researchers in the field of breast cancer classification to develop more effective and reliable studies.
Turkish Named Entity Discovery Based on Termsets
(IEEE, 2019) Coban, Onder; Ozel, Selma Ayse; Iean, Ali
Named Entity Recognition (NER) is a subtask of the information extraction process and aims to discover named entities in unstructured texts. Previous studies on NER mostly use statistical machine learning models instead of using classifiers since solving this problem as a classification task requires to deal with quite high dimensional and sparse vector spaces. In this paper, we take NER as a classical text classification problem and extract nominal features from each token in the unstructured text sequence. We convert each token to a document transaction and then, we use frequent termset mining to extract termset features and apply termset weighting to classify named entities. Therefore we deal with lower dimensional feature spaces. Our experimental results obtained on a large Turkish dataset show that frequent termsets and their weighting scheme can be used in NER task.
Using BERT models for breast cancer diagnosis from Turkish radiology reports
(Springer, 2024) Hepsag, Pinar Uskaner; Ozel, Selma Ayse; Dalci, Kubilay; Yazici, Adnan
Diagnostic radiology is concerned with obtaining images of the internal organs using radiological imaging procedures. These images are then interpreted by a diagnostic radiologist, who produces a textual report that assists in the diagnosis of illness or injury. Early detection of certain illnesses, particularly cancer, is critical, and the reports produced by diagnostic radiologists play a key role in this process. To develop models for the early detection of cancer, text classification techniques can be applied to radiological reports. However, this process requires access to a dataset of radiology reports, which is not widely available. Currently, radiology report datasets exist for high-resource languages such as English and Dutch, but not for low-resource languages such as Turkish. This article describes the collection of a mammography report dataset for Turkish, consisting of 62 reports from real patients that were manually labeled by an expert for diagnosing breast cancer. Basic machine learning models were applied to this dataset using pre-trained BERT, DistilBERT, and an ensemble learning hard voting approach. The results showed that BERT on Turkish achieved the best performance, with a 91% F1-score. Hard Voting, which combined the results of BERTTurkish, BERTClinical, and BERTMultilingual, achieved the highest F1-score of 93%. The results show that BERT and Hard Voting outperform the other machine learning models for breast cancer diagnosis from Turkish radiology reports.
Using Fuzzy Sets for Detecting Cyber Terrorism and Extremism in the Text
(IEEE, 2018) Uzel, Vahide Nida; Essiz, Esra Sarac; Ozel, Selma Ayse
The concept of Cyber Security (CS) has been started to be used with the development of Internet technology. Nowadays, CS has vital importance and Cyber Terror and Extremism (CTE) is one of the CS problems. Terror must be detected before terrorism comes true. In other words, people who commit the crime must be detected automatically before they move on. At this stage, what people say about some issues is very valuable because sayings can be turned into actions. The aim of this study is to use Antisocial Behavior dataset to try to detect CTE in the text contents. To detect CTE, text documents should be converted to numerical vectors which consist of numerical weights of the terms present in the text documents. Vectors are computed by using four different weighting methods in our study. These methods are the well-known binary weighting, term frequency based weighting, term frequency and inverse document frequency based weighting, and our proposed fuzzy set based weighting methods. Naive Bayes Multinomial (NBM) and Support Vector Machines (SVM) are used as classifiers to compare the performances of the weighting methods for CTE detection. Our experimental analysis shows that fuzzy set based weighting method with SVM classifier gives the best classification accuracy which reaches up to 99%.

Yazar "Ozel, Selma Ayse" seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri