TOBB ETU GCRIS Database
Not a member yet
9778 research outputs found
Sort by
Prediction of Migration Paths Using Agent-Based Simulation Modeling: the Case of Syria
No full textThe Syrian civil war, which started in 2011, has caused a great wave of forced migration in the Middle East. One of the most popular destination points for Syrian refugees has been Turkey. The purpose of this study is to predict the routes of refugees who leave the conflict areas in Syria to reach the refugee camps located in Turkey during a crisis. The study proposes an agent-based model to simulate the decision mechanisms of refugees in a highly uncertain environment. The model employs the A* algorithm to calculate the cost of each available destination point (refugee camp) for each agent, based on their risk preferences and starting locations, and allows agents to choose the camp with the minimum cost as the destination point. By use of the model, we simulate a moment of crisis namely the South Idlib bombardment (from December 2019 to January 2020) under four different scenarios that are generated considering the real-life data gathered from the newspapers of December 2019 and various other sources. The simulation results show the main pathways of Syrian refugees and give insights on the required camp capacities. The results are compared with the gathered secondary data to validate the proposed model
Transmission of Cholera Disease With Laplacian and Triangular Parameters
No full textA mathematical model has been introduced for the transmission dynamics of cholera disease by GQ Sun et al. recently. In this study, we add Laplacian and Triangular random effects to this model and analyze the variation of results for both cases. The expectations and co-efficients of variation are compared for the random models and the results are used to comment on the differences and similarities between the effects of these probability distributions. The randomness of the model itself is also investigated through comparison of the random and deterministic outcomes. © 2022 Academic Center for Education, Culture and Research TMU
Uluslararası Özel Hukuk Perspektifinden Akıllı Sözleşmeler
No full textBlokzincir teknolojisinin akıllı sözleşmeler için de kullanılmaya başlanması, hukukun farklı alanlarını etkileyecek önemli bir gelişmedir. Şeffaf, güvenilir, hızlı ve daha az masraflı olarak nitelendirilen akıllı sözleşmelerin uyandırdığı ilgi karşısında mevcut hukuk kurallarının bu sözleşmelerle bağlantılı ihtiyaçlara cevap verip veremeyeceklerinin belirlenmesi elzem hale gelmiştir. Bu kapsamda yapılan çalışmalar, genellikle, sözleşmeler hukuku prensipleri çerçevesinde akıllı sözleşmelerin hukuki niteliği konusunu ele almaktadır. Ancak, blokzincir teknolojisinin dağıtık yapısı nedeniyle bunlar üzerinde kurgulanan akıllı sözleşmeler, kanunlar ihtilafı ile milletlerarası usul hukuku açısından da incelenmeye muhtaçtır. Bu makalede, karşılaştırmalı hukukta akıllı sözleşmeleri hukuken geçerli ve bağlayıcı bir sözleşme olarak kabul etme doğrultusundaki genel eğilim esas alınarak uygulanacak hukuk ve uyuşmazlık çözümü konuları mevcut hukuki altyapı çerçevesinde değerlendirilmiştir.The use of blockchain technology also for smart contracts is an important progress which will affect different areas of law. Smart contracts attract attention with their transparent, reliable, fast and less expensive structure and it is now essential to specify whether existing law rules can meet the needs connected to these contracts. The studies made in this context, generally refer to the legal status of smart contracts within the principles of contract law. However, smart contracts based on blockhains need to be examined also from the conflict of laws and international procedural law perspectives because of the distributed structure of blockchain technology. This paper, leaning on the general tendency in comparative law to accept smart contracts as legally valid and binding, evaluates the issues of applicable law and dispute resolution according to the existing legal framework
Age and Gender Classification From Facial Features and Object Detection With Machine Learning
No full textIn recent years, development of the machine learning algorithms has led to the creation of intelligent surveillance systems. Thanks to the machine learning, it is possible to perform intelligent surveillance by recognizing people's facial features, classifying their age and gender, and detecting objects around instead of ordinary surveillance. In this study, a novel algorithm has been developed that classifies people's age and gender with a high accuracy rate. In addition, a novel object recognition algorithm has been developed that detects objects quickly and with high accuracy. In this study, age and gender classification was made based on the facial features of people using Convolutional Neural Network (CNN) architecture. Secondly, object detection was performed using different machine learning algorithms and the performance of the different machine learning algorithms was compared in terms of median average precision and inference time. The accuracy of the age and gender classification algorithm was tested using the Adience dataset and the results were graphed. The experimental results show that age and gender classification algorithms successfully classify people's age and gender. Then, the performances of object detection algorithms were tested using the COCO dataset and the results were presented in graphics. The experimental results stress that machine learning algorithms can successfully detect objects. © 2022, Research Expansion Alliance (REA). All rights reserved
Overview of the Clef-2022 Checkthat! Lab Task 1 on Identifying Relevant Claims in Tweets
No full text2022 Conference and Labs of the Evaluation Forum, CLEF 2022 -- 5 September 2022 through 8 September 2022 -- -- 181762We present an overview of CheckThat! lab 2022 Task 1, part of the 2022 Conference and Labs of the Evaluation Forum (CLEF). Task 1 asked to predict which posts in a Twitter stream are worth fact-checking, focusing on COVID-19 and politics in six languages: Arabic, Bulgarian, Dutch, English, Spanish, and Turkish. A total of 19 teams participated and most submissions managed to achieve sizable improvements over the baselines using Transformer-based models such as BERT and GPT-3. Across the four subtasks, approaches that targetted multiple languages (be it individually or in conjunction, in general obtained the best performance. We describe the dataset and the task setup, including the evaluation settings, and we give a brief overview of the participating systems. As usual in the CheckThat! lab, we release to the research community all datasets from the lab as well as the evaluation scripts, which should enable further research on finding relevant tweets that can help different stakeholders such as fact-checkers, journalists, and policymakers. © 2022 Copyright for this paper by its authors.Hamad Bin Khalifa University, HBKUPart of this work is made within the Tanbih mega-project, developed at the Qatar Computing Research Institute, HBKU, which aims to limit the impact of “fake news”, propaganda, and media bias by making users aware of what they are reading, thus promoting media literacy and critical thinking
The Clef-2022 Checkthat! Lab on Fighting the Covid-19 Infodemic and Fake News Detection
No full text44th European Conference on Information Retrieval (ECIR) -- APR 10-14, 2022 -- Stavanger, NORWAYThe fifth edition of the CheckThat! Lab is held as part of the 2022 Conference and Labs of the Evaluation Forum (CLEF). The lab evaluates technology supporting various factuality tasks in seven languages: Arabic, Bulgarian, Dutch, English, German, Spanish, and Turkish. Task 1 focuses on disinformation related to the ongoing COVID-19 infodemic and politics, and asks to predict whether a tweet is worth fact-checking, contains a verifiable factual claim, is harmful to the society, or is of interest to policy makers and why. Task 2 asks to retrieve claims that have been previously fact-checked and that could be useful to verify the claim in a tweet. Task 3 is to predict the veracity of a news article. Tasks 1 and 3 are classification problems, while Task 2 is a ranking one.This research is part of the Tanbih mega-project, developed at the Qatar Computing Research Institute, HBKU, which aims to limit the impact of fake news, propaganda, and media bias, thus promoting media literacy and critical thinking. The Arabic annotation effort was partially made possible by NPRP grant NPRP13S-0206-200281 from the Qatar National Research Fund (a member of Qatar Foundation).Univ Stavanger, British Comp Soc, Informat Retrieval Specialist Grp, Amazon, Bloomberg, Cobrainer, Elsevier, Google, L3S Res Ctr, MediaFutures, Norwegian Univ Sci ; Technol, NorwAI, Schibsted, SIGIR, Signal AI, Spotify, Springer, Textkernel, Thomson Reuters, Vespa AI, Wayfair, Norwegian Univ Sci ; Technol, Dept Comp SciNPRP from the Qatar National Research Fund (a member of Qatar Foundation) [NPRP13S-0206-200281
Yogun Bakim Hastalarinin Mortalite ve Hastanede Kalma Sürelerinin Derin Ögrenme Yöntemleri ile Tahmini
No full textThe increase in the speed of digital transformation has accelerated the transfer of physically stored data to electronic media data. Healthcare is one of many areas that is impacted by these transformation. Electronic health records (EHR) is the general term for the data that is associated with a patient's whole health journey including demographic information, laboratory test results, vital signs, clinical notes, diagnosis codes, and related data. In addition to the digital transformation in healthcare, the widespread interest in machine/deep learning encourages the researchers to apply artificial intelligence to several different domains such as finance, social media, and cyber security. With the EHR data becoming available for researchers, there has been an increasing interest in using it with deep learning algorithms. Within the scope of this study, we use the most popular and publicly available EHR dataset, Medical Information Mart for Intensive Care (MIMIC-III). Understanding the health condition of the patient by observing the clinical measurements, and laboratory tests, and predicting the condition of patients during their intensive care unit (ICU) stay is a vital problem. In this study, two different common risk prediction tasks, mortality (in-hospital \; in-ICU), and length of ICU stay (LOS 3, LOS ) are researched. The interest of this work is divided into three parts. In the first part, we use the clinical notes besides the time-series features such as vital signs and laboratory test results to improve the model predictions. In the second part, instead of using clinical notes directly, we extract medical entities from clinical notes by clinical named entity recognition (NER) model and use them as additional features besides time-series features to improve proposed model predictions. In the last study, we argue the integration of structured time-series data and molecular representations of the drugs which are prescribed to patients in ICU. Several experiments are conducted to investigate the effect of clinical drugs on mortality and LOS problem predictions. Additionally, the SHapley Additive exPlanations (SHAP) is applied to increase the interpretability of the in-hospital mortality model and to investigate the relationship between the mortality and the time-series and clinical drug features. The output of the SHAP method allows us to make a deeper analysis of the effect of time-series and clinical drug features. Since MIMIC-III contains rich information with multiple modalities of data, we apply a multimodal learning approach to handle the heterogeneous nature of the data. The experimental results indicate a promising increase in performance on clinical tasks when the clinical notes, medical entities or clinical drug informations are used with time series features in a multimodal approach.Günümüzde dijital dönüşüm hızının artması ile beraber fiziksel olarak saklanan verilerin elektronik ortamlara aktarılması da hız kazanmıştır. Bu durum, birçok alana olduğu gibi sağlık alanına da doğrudan yansımıştır. Geçmişte fiziksel olarak saklanan hastaya ait kayıtlar bu sayede dijital ortamlara geçirilmiştir. Dijital ortama aktarılan hastaya ait demografik bilgiler, laboratuvar sonuçları, yaşamsal gözlem verileri, klinik notlar, tanı kodları ve benzeri birçok veri Elektronik Sağlık Kaydı (ESK) olarak tanımlanmaktadır. Sağlık alanındaki dijital dönüşüme ek olarak, derin öğrenme yöntemlerine olan geniş ilgi, araştırmacıları, finans, sosyal medya, siber güvenlik gibi birçok alanda yapay zeka yöntemlerini kullanmaya teşvik etmektedir. Elektronik sağlık kayıtlarının araştırmacılar için kullanılabilir hale gelmesiyle birlikte, bu veri setlerini kullanarak derin öğrenme modelleri geliştirmeye olan ilgi artmaktadır. Tez kapsamında yapılan deneylerde, günümüzdeki en popüler ve erişilebilir elektronik sağlık kayıt veri seti olan Medical Information Mart for Intensive Care (MIMIC-III) kullanılmıştır. Yoğun bakımda yatan hastaların, yaşamsal gözlem verilerini ve diğer klinik bilgilerini ölçerek, hastaların mevcut sağlık durumlarını anlamlandırmak ve gelecek sağlık durumlarını tahmin etmek önemli bir problemdir. Tez kapsamında, hastaların hastane içinde ve yoğun bakımda mortalite ihtimalleri ile yoğun bakımda 3 ve 7 günden fazla kalıp kalmayacakları çok-kipli derin öğrenme tabanlı yöntemler ile tahmin edilmiştir. Gerçekleştirilen çalışma üç ana bölüme ayrılmıştır. İlk bölümde, yoğun bakımda yatan hastalara ait yaşamsal gözlem verileri, laboratuvar sonuçları gibi özniteliklere ek olarak hastalara ait klinik notlar da model eğitimine dahil edilmiş ve modelin klinik problemleri tahmin etme başarısı arttırılmaya çalışılmıştır. İkinci bölümde, klinik notların doğrudan kullanılması yerine, varlık isim tanıma yöntemi ile notlar içerisinden medikal terimlerin çıkartılması sağlanmıştır. Elde edilen medikal terimlerin, mortalite ve yoğun bakımda kalma süresi tahmini problemlerine etkisi araştırılmıştır. Yapılan son çalışmada ise, hastaların zaman serisi özniteliklerine ilave olarak, hastaların yoğun bakımda kaldıkları süre boyunca kullandıkları ilaçların moleküler temsilleri kullanılmış ve klinik problemlerin tahminine etkisi üzerine deneyler yapılmıştır. Ek olarak, bu çalışma sonunda, hastanede mortalite tahmini için eğitilen modelin açıklanabilirliğini arttırmak amacıyla SHapley Additive exPlanations (SHAP) yöntemi kullanılmıştır. SHAP yönteminin çıktısı, zaman-serisi ve klinik ilaç özniteliklerinin model üzerindeki etkisininin daha derin bir analizinin yapılmasını sağlamaktadır. MIMIC-III veri seti içerisinde hastaya ait farklı veri türlerinin bir arada bulunması, tez kapsamında yapılan deneylerde bu veri türlerinin bir arada kullanılabilmesine ve farklı deneylerin gerçekleştirilebilmesine olanak sağlamıştır. Farklı veri türlerini aynı model içerisinde kullanabilmek için çok-kipli derin öğrenme tabanlı yöntemler önerilmiştir. Yapılan deney sonuçları incelendiğinde, zaman-serisi özniteliklerin yanı sıra hastaya ait klinik notların, medikal terimlerin ve ilaç bilgilerinin modele girdi olarak verilmesinin, klinik problemlerin başarımına olumlu yönde etki ettiği görülmüştür
Türkiye'de Ekonomi Yöneticilerinin Değişiminin Borsa Endekslerine Olan Etkisi: Bist Endekslerinde Bir Uygulama
No full textThe aim of this study is to examine the effects of the governor changes the Central Bank of the Republic of Turkey and the Ministry of Treasury and Finance of the Republic of Turkey on six different stock market indices. While examining this effect, the last three changes that took place in both institutions were taken into account. In the study, daily returns of BIST100, BIST BANKA, BIST GIDA, BIST TURİZM, BIST TİCARET and BIST INDUSTRY indices between 2018 and 2021 were examined by event analysis method. The t-test was used while performing the analysis. As a result of the study, it has been determined that the governor changes have varying effects on the abnormal returns of the six different indices.Bu çalışmanın amacı, Türkiye Cumhuriyet Merkez Bankası Başkanlığında ve Türkiye Cumhuriyeti Hazine ve Maliye Bakanlığında gerçekleşmiş olan yeni atamaların belirlenmiş olan altı farklı borsa endeksine olan etkisini incelemektir. Bu etki incelenirken iki kurumda da gerçekleşmiş olan son üç değişim dikkate alınmıştır. Çalışmada BİST100, BİST BANKA, BİST GIDA, BİST TURİZM, BİST TİCARET ve BİST SANAYİ endekslerinin 2018 ile 2021 yılları arasındaki günlük getirileri olay çalışması yöntemi ile inceleme altına alınmıştır. Analiz gerçekleştirilirken t-testi kullanılmıştır. Çalışmanın sonucunda, gerçekleşmiş olan değişimlerin belirtilen altı farklı endeksin anormal getirileri üzerinde çeşitlilik gösteren farklı etkilere sahip olduğu saptanmıştır
Pandemide Etik Sorunlar Nasıl Ele Alınmalı? Yeni Bir Kuramsal Çerçeve Önerisi
No full textPandemi, insanlığın varlığını ve iyilik halini tümden tehdit eden bir küresel bir halk sağlığı acil durumudur. Halk sağlığı alanındaki etik sorunlara yönelik geliştirilmiş çözümleme metodolojileri bölgesel ve sınırlı halk sağlığı acillerinde bir miktar yol gösterici olabilse de, pandemi gibi küresel acil durumlarda yeterli rehberlik sağlayamamaktadır. COVID-19 pandemisi bu eksikliği belirgin bir şekilde görmemizi sağlamıştır. Bu makalede öncelikle halk sağlığı etiğinin (HSE) temel unsurları, sorun alanları ve çözüm yaklaşımlarına ilişkin bir çerçeve sunulacak, daha sonra bu çerçevenin küresel halk sağlığı acillerinde neden yeterli rehberlik sağlamadığı açıklanacak ve son olarak yeni bir etik çözümleme çerçevesi önerisinde bulunulacaktır.The pandemic is a global public health emergency that threatens the existence and wellbeing of humanity as a whole. While improved resolution methodologies for ethical issues in public health may provide some guidance in regional and limited public health emergencies, they do not provide adequate guidance in global emergencies such as pandemics. The COVID-19 pandemic has made this deficiency clear. In this article, firstly, a framework regarding the basic elements of public health ethics (HSE), problem areas and solution approaches will be presented, then it will be explained why this framework does not provide sufficient guidance in global public health emergencies, and finally, a new ethical analysis frame will be proposed
End-tidal carbon dioxide levels under surgical drapes during local eye surgery: Retrospective study
No full textObjectives: To investigate the end tidal carbon dioxide pressure values in order to determine carbon dioxide accumulation under surgical drapes and it’s hemodynamic effects based on anesthetic and surgical records in eye surgeries under local anesthesia Design: Retrospective study Setting: Department of Anesthesiology and Reanimation, Düzce Medical Faculty, Düzce, Turkey Subject: The data were collected from anesthetic records of patients (n=42) who were followed with noninvasive capnography in the operating room at Düzce University Hospital during the period of January 2016 to December 2016. Systolic, diastolic and mean arterial pressure, operation time, heart rate, ST segment analysis, ETCO2 pressure, pulse oximetry values ??were recorded. Time periods were determined as: after the anesthesia and before drape closure (baseline level), at 10t h, 15th, 20th, 45th of the surgery and 5 minutes after drape removal. Intervention: Non-interventional Main outcome measure: C a rbon dioxide accumulation under drapes and it’s hemodynamic effects in eye surgeries under local anesthesia. Results: The comparisons were made with basal status and time periods statistically. No differences were found between mean arterial pressures, heart rates, arrhythmias and pulse oximetry values of pati ents between time periods. We observed significant di fferences for ETCO2 levels between basal and the other ti me periods, except when the drapes were removed (P=0.001). Conclusion: Routine mo nitorisation of ETCO2 with noninvasive capnograph yprovides early detection of CO2 accumulation and CO2 rich air breathing during ophthalmic surgery. © 2022, Kuwait Medical Association. All rights reserved