Mühendislik Fakültesi, Bilgisayar Mühendisliği (İngilizce Destekli) Bölümü Koleksiyonu

Bu koleksiyon için kalıcı URI

https://hdl.handle.net/11467/9346

Listeleniyor 1 - 20 / 60

Sohbet robotu ile kullanıcı yönetimi
(İstanbul Ticaret Üniversitesi, 2025) Bulut, Muhammed Burak; Aksoy, Saadettin
Bir bayi ağı ve bu ağın kuryeleri tarafından kullanılan CRM sisteminde kullanıcı adı ve şifre yönetimini iyileştirmek için geliştirilmiş bir chatbot sistemini ele almaktadır. Amaç, kuryelere güvenli ve verimli bir şekilde kullanıcı adı ve şifre sağlanmasını otomatikleştirmektir. Sistem, kuryelerin telefon numaralarını analiz ederek doğru kullanıcı adı ve şifreleri oluşturur ve bunları chatbot aracılığıyla iletir. Bu süreçte, kullanıcı adı ve şifrelerin güvenli bir şekilde oluşturulması ve iletilmesi detaylandırılmıştır. Ayrıca, veriseti oluşturulma süreci, kullanılan yazılım araçları ve ortamı açıklanmıştır. Deneysel sonuçlar, chatbotun yüksek doğruluk ve güvenlik sağladığını, müşteri memnuniyetini artırdığını ve şirket içi iş süreçlerini verimli hale getirdiğini göstermektedir. Gelecekte, sistemin farklı güvenlik önlemleri ile geliştirilmesi ve diğer uygulama alanlarına adaptasyonu önerilmektedir.
Dalgaboyu bölümlemeli çoğullama (DBÇ) üzerinde optik çoğuşma anahtarlama (OÇA) ve Çoğa Gönderim
(2015) Kırcı, Pınar; Zaim, Halim
Son yıllarda internet kullanımının katlanarak artması, kullanıcılara sunulan internet uygulamalarındaki çeşitlilik ve bunların hızlı gelişimi, çok büyük oranlarda bantgenişliği talebini ortaya çıkarmıştır. Özellikle bire-çok bağıntı ya da çoktan-çok bağıntı iletişimi yapısında olan çoğa gönderim uygulamalarını temel alan yüksek bantgenişliği gerektiren uygulamalardan video konferans, uzaktan interaktif eğitim gibi örnekler optik ağlarda Dalgaboyu Bölümlemeli Çoğullama-DBÇ (Wavelength Division Multiplexing- WDM) teknolojisi sayesinde kullanıcılara kolaylıkla ulaştırılmaktadır. Bu çalışmada Optik Çoğuşma Anahtarlama (Optical Burst Switching-OBS) ağlarda çoğa gönderimi sağlayabilmek için Önceden Hesaplanan Yol Çoğa Gönderim (Precalculated Path Multicast-PPM) isimli yeni bir protokol sunulmuştur. Protokol yapısı ve işleyişi durum diyagramları kullanılarak açıklanmıştır.
The proposal of a fuzzy AHP-based model for the efficiency analysis of technoparks
(2019) Tepe, Serap; Zaim, Abdul Halim
In the paper, it is proposed a model to evaluate the efficiency of technopark by performing an activity analysis on the productivity of the technopark structure, aiming to contribute to the innovative movement and sustainable development goals, which may result in huge amounts of increase in added-value of Turkey’s technological development. The initial point of the paper is to define technopark structure and activities, to determine and eliminate uncertainties concerning structure and operation. Technoparks are investigated based upon their management, firms & incubation firms, R&D activities, and cooperation level among university-industry. In the paper, a fuzzy analytic hierarchy process method is used. The paper includes four technoparks that operate in Istanbul. By applying the developed model, the results of the performance evaluation are reached and the results are interpreted. It is thought that the findings obtained from this research will be beneficial for all stakeholders related to technoparks.
Uncovering the dynamics in the application of machine learning in computational finance: A bibliometric and social network analysis
(Econjournals, 2024) Ajibade, Samuel-Soma Mofoluwa; Jasser, Muhammed Basheer; Alebiosu, David Olayemi; Al-Hadi, Ismail Ahmed Al-Qasem; Al-Dharhani, Ghassan Saleh; Hassan, Farrukh; Gyamfi, Bright Akwasi
This paper examined the research landscape on the applications of machine learning in finance (MLF) research based on the published documents on the topic indexed in the Scopus database from 2007 to 2021. Consequently, the publication trends on the published documents data were examined to determine the most prolific authors, institutions, countries, and funding bodies on the topic. Next, bibliometric analysis (BA) was employed to analyse and map co-authorship networks, keywords occurrences, and citations. Lastly, a systematic literature review was carried out to examine the scientific and technological developments in the field. The results showed that the number of published documents on MLF research has soared tremendously from 5 to 398 between 2007 and 2021, which signifies an enormous increase (~7,900%) in the subject area. The high productivity is partly ascribed to the research activities of the most research-active academic stakeholders namely Chihfong Tsai (National Central University in Taiwan) and Stanford University (United States). However, the National Natural Science Foundation of China (NSFC) is the most active funder in the United States and has the largest number of published documents. BA analysis revealed high collaboration rates, published documents, and citations among the stakeholders. Keywords occurrence analysis revealed that MLF research is a highly inter-and multidisciplinary area with numerous hotspots and themes ranging from systems, algorithms and techniques to the security and crime prevention in Finance using ML. Citation analysis, the most prominent (and by extension the most prestigious) source titles on MLF are IEEE Access, Expert Systems with Applications and ACM International Conference Proceedings Series (ACM-ICPS). The systematic literature review revealed the various areas and applications of MLF research, particularly in the areas of predictive/ forecasting analytics, credit assessment and management, as well as supply chain, carbon trading, neural networks, and artificial intelligence, among others. It is expected that MLF research activities and their impact on the wider global society will continue to increase in the coming years.
Enhanced chaotic key-based algorithm for low-entropy image encryption
(IEEE Computer Society, 2014) Yavuz, Erdem; Yazıcı, Rıfat; Kasapbaşı, Mustafa Cem; Yamaç, Ezgi
As computing power of today's computer enhances rapidly, secure encryption is getting harder and the need for powerful secure encryption methods becomes inevitable. To avoid brute-force attacks it is necessary to have a larger key space, which makes the chaotic systems more attractive. In order to sufficiently apply confusion and diffusion rules for an image data with any entropy structure, a two-dimensional chaotic system, which has high sensitivity to initial states, has been utilized in this study. The fact that adjacent pixels of an image have naturally close values makes it easier for crackers to perform statistical analysis. So, two chaotic functions are used to disrupt correlations among them, one for shuffling pixel positions and one for changing pixels values. Even with low entropy images the proposed algorithm has been proved to be more secure and faster than the previous algorithms.
Bibliographic exploration of application of machine learning and artificial intelligence in solar energy
(Institute of Electrical and Electronics Engineers Inc., 2024) Ajibade, Samuel-Soma Mofoluwa; Mohammed Bashir, Faizah ; Dodo, Yakubu Aminu; Oyebode, Oluwadare Joshua; Culpable, Rex V.; De La Calzada, Limic M.; Adediran, Anthonia Oluwatosin; Ajibade, Temiloluwa Iyanuoluwa
Solar energy could mitigate global warming and climate change. Solar energy faces economic, environmental, and technical challenges. Machine learning solves these technical issues. Despite several studies, machine learning in photovoltaics and solar energy is understudied. This study examines publishing patterns and bibliometrics to critically evaluate machine learning applications in photovoltaics and solar energy research. Scopus uses PRISMA. International publishing, citations, and collaboration are high. The Chinese Ministry of Education employs famous scholars like G. E. Georghiou and Haibo Ma. China is most active due to funding schemes like the National Natural Science Foundation and the National Key Research and Development Programme. This study examines publication patterns by country, institution, and funding organisation from 2014 to 2022, spanning topic categories and indicators. Examining author-keyword data to group publishing themes and identify influential journals. Increasing understanding of machine learning applications in photovoltaics and solar energy research. This project will examine the potential for significant development and the hurdles that must be overcome to leverage Cognitive Computing's benefits in cancer and tumour research. In response to the rising amount of malware, phishing, and intrusion attacks on global energy and grid infrastructure, photovoltaic and solar energy system cybersecurity may be studied. © 2024 IEEE.
Application of artificial intelligence in healthcare systems: A scientometric analysis
(Institute of Electrical and Electronics Engineers Inc., 2024) Ajibade, Samuel-Soma Mofoluwa; Lee, Angela Siew Hoong; Jasser, Muhammed Basheer; Akinola, Tomiwa Faith
This bibliometric study examines the worldwide progression of machine learning applications in medical and healthcare research between 2010 and 2023. A large dataset of published articles pertaining to machine learning applications in the medical and healthcare sectors is mined for useful information in this study. The data extraction procedure involves retrieving pertinent information from primary sources such as journals, books, and conference proceedings. Subsequently, the retrieved data is subjected to analysis to discern the patterns and tendencies in the use of machine learning in medical and healthcare research. A total of 1,220 publications were found in the Scopus database over the past 14 years. In addition, the study demonstrated that most AIHS research has concentrated on artificial intelligence applications to address a wide range of problems, including patient data security and chronic medical difficulties (such as cardiovascular disorders). Policymakers, healthcare practitioners, and researchers around the world may find this study's conclusions helpful. Emerging ethical concerns, integration, and real-world uses in smart healthcare systems, and the Internet of Things (IoT), could be the focus of future research.
The capacitance/conductance and surface state intensity characteristics of the Schottky structures with ruthenium dioxide-doped organic polymer interface
(Elsevier Science Sa, 2023) Ulusoy, Murat; Badali, Yosef; Pirgholi-Givi, Gholamreza; Azizian-Kalandaragh, Yashar; Altindal, Semsettin
The electrical behaviors of the Schottky structures with a ruthenium dioxide (RuO2) doped-polyvinyl chloride (PVC) interface were executed with a wide frequency range (from 1 kHz to 5 MHz) and voltages. The interface was obtained by dispersing RuO2 nanopowder as colloidal particles into the PVC organic polymer using the ultrasonic irradiation method. The capacitance/conductance and surface state intensity (Nss) effects of this interface on the structure have been widely discussed. Remarkable increases in capacitance (C) and conductance (G/.) values were found, especially in the depletion zone. The series resistance of the structure (Rs) value decreases strongly with increasing frequency for + 3.5 V, down to a value of approximately 48.43 O at 5 MHz. Furthermore, the effect of the Rs is seen in the Cc and Gc/. curves in the weak and strong accumulation regions. While the maximum value of the Nss is 1.42 x 10(13) eV 1.cm(-2) at 0.478 eV, its minimum value is 1.23 x 1013 eV 1.cm (-) at 0.540 eV. The relaxation time (t) values change from 2.40 x 10(-5) to 2.03 x 10(-4) s in exponentially increasing values. It can be stated that there is an inverse relationship between the t and distribution of the Nss values. These distributions vary depending on the applied voltage and frequency.
The temperature-dependent dielectric properties of the Au/ZnO-PVA/n-Si structure
(Elsevier, 2023) Azizian-Kalandaragh, Yashar; Badali, Yosef; Jamshidi-Ghozlu, Mir-Ahmad; Hanife, Ferhat; Özçelik, Süleyman; Altındal, Şemsettin; Pirgholi-Givi, Gholamreza
In this work, the temperature-dependent (80-360 K) dielectric properties of the Au/ZnO-PVA/n-Si structure was investigated employing capacitance-voltage (C-V) and conductance-voltage (G/ro-V) experiments at 1 MHz. The results indicate that all electrical and dielectric variables in these structures are forcefully dependent on tem-perature. Also, using the interlayer ZnO-PVA nanocomposite has caused changes to these parameters. Because of the presence of series resistance, the amount of C and G/ro increases as the temperature rises. The values of EF increase with temperature, whereas the values of barrier height decrease from 1.045 eV to 0.943 eV, and the value of alpha extract from phi B-T plot is obtained-3.5 x 10-4 eV/K that is approximately equal to the silicon temperature coefficient. The value of activation energy is obtained 0.04 eV which is a modest amount obtained from the conduction procedure's contribution to the boundary grains.
A systematic review on software reliability prediction via swarm intelligence algorithms
(Elsevier, 2024) Kong, Li Sheng; Jasser, Muhammed Basheer; Ajibade, Samuel-Soma Mofoluwa; Mohamed, Ali Wagdy
The widespread integration of software into all parts of our lives has led to the need for software of higher reliability. Ensuring reliable software usually necessitates some form of formal methods in the early stages of the development process which requires strenuous effort. Hence, researchers in the field of software reliability introduced Software Reliability Growth Models (SRGMs) as a relatively inexpensive approach to software reliability prediction. Conventional parameter estimation methods of SRGMs were ineffective and left more to be desired. Consequently, researchers sought out swarm intelligence to combat its flaws, resulting in significant improvements. While similar surveys exist within the domain, the surveys are broader in scope and do not cover many swarm intelligence algorithms. Moreover, the broader scope has resulted in the occasional omission of information regarding the design for reliability predictions. A more comprehensive survey containing 38 studies and 18 different swarm intelligence algorithms in the domain is presented. Each design proposed by the studies was systematically analyzed where relevant information including the measures used, datasets used, SRGMs used, and the effectiveness of each proposed design, were extracted and organized into tables and taxonomies to be able to identify the current trends within the domain. Some notable findings include the distance-based approach providing a high prediction accuracy and an increasing trend in hybridized variants of swarm intelligence algorithms designs to predict software reliability. Future researchers are encouraged to include Mean Square Error (MSE) or Root MSE as the measures offer the largest sample size for comparison.
Application of carbon-based nanomaterials in solar-thermal systems: An updated review
(Pergamon-Elsevier Science Ltd, 2024) Tuncer, Azim Doğuş; Badali, Yosef; Khanlari, Ataollah
Increasing the world population makes it necessary to look for alternative and sustainable energy resources to meet the rising energy demand. In recent years, the utilization of different renewable energies has been accelerated because of the negative effects of fossil fuels and their limited resources. Among different solar energy technologies, solar liquid and air collectors are extensively utilized for providing heated water and air. Rapid development in science and technology makes it possible to improve the performance of solar collectors. Producing nanomaterials and using the obtained nano-sized particles for improving the thermophysical specifications of the working fluid or coating material can be considered for performance improvement. In the present study, applications of carbon-based nanomaterials (CBNMs) in various solar thermal systems have been reviewed comprehensively. In other words, the effects of utilizing carbon-based nanomaterials as promising material on the overall performance have been investigated. Preparation techniques and thermophysical specifications of carbon-based nanomaterials have been explained in detail. Also, utilization of carbon-based nanomaterials in seven widely utilized solar-thermal technologies and their impacts on the thermal behavior of the tested system have been concluded. This review gives a general perspective about applicability of carbon-based nanomaterials for performance enhancement in various solar-thermal systems.
An acoustic signal based language independent lip synchronization method and its implementation via extended LPC
(IEEE, 2020) Cankurtaran, Halil Said; Boyacı, Ali; Yarkan, Serhan
Processing human speech with the use of digital technologies leads to several important fields of research. Speech-to-text and lip-syncing are among the instances of relevant prominent research areas. In this regard, audio-visualization of acoustic signals, providing visual aid in real-time for disabled people, and realization of text-free animation applications are just to name a few. Therefore, in this study, a language-independent lip-sync method that is based on extended linear predictive coding is proposed. The proposed method operates on baseband electrical signal that is acquired by a standard single-channel off-the-shelf microphone and exploits the statistical characteristics of acoustic signals produced by human speech. In addition, the proposed method is implemented on an embedded system, tested, and its performance is evaluated. Results are given along with discussions and future directions.
Blok zinciri platformları, fikir birliği mekanizmaları ve ağın güvenlik analizi
(Haliç Üniversitesi, 2022) Aslan, Mimar; Kasapbasi, Mustafa Cem
Finans, sağlık, sosyal medya vb ortamlardaki insanların ihtiyacı olan güven problemine blok zinciri teknolojisi, şifreli algoritmalar ile çözümler sunmaktadır. Güven problemini çözen ve verileri dağıtık olarak kayıt altına alan ve her şeyi şeffaf olarak bizlere sunan blok zinciri bir devrim niteliğindedir. Blok zinciri, akıllı sözleşmeler sayesinde kurumsal projelerde de kullanılmaktadır. Kurumların arasında yeni nesil bir ağ olarak da adlandırılan blok zinciri ile birçok şeyin değişmesi beklenmektedir. İnternetin, mobil cihazların ve sensörlerin yaygınlaşmasıyla birlikte güven problemi her geçen gün daha da önem kazanmaktadır. Farklı amaçlara hizmet eden Ethereum, Cardano, EOS, Cosmos, Hyperledger gibi blok zincir platformları bulunmaktadır. Altyapılarında Proof of Work (PoW), Proof of Stake (PoS), Delegated Proof of Stake (DPoS) ve Directed Acyclic Graph (DAG) gibi farklı fikir birliği mekanizmalarını kullanmaktadırlar. Bu çalışmada blok zinciri platformları, altyapılarında kullandıkları mutabakat mekanizmaları ve blok zinciri ağının güvenliği incelenrek araştırılmıştır.
"5G: Next generation mobile wireless technology for a fast pacing world"
(Journal of Pure and Applied Sciences, 2022) Ajibadae, Samuel Mofoluwa
The quest for faster and seamless mobile communication services has given rise to the Fifth Generation (5G) mobile communication technology. 5G presents enormous opportunities for endusers, businesses, and the global economy at large. The major edge 5G has over its predecessor technologies; the 3rd and 4th Generation (3G and 4G), is ultra-low latency, which translates to ultrafast data speed, which has the potential to peak to about 1Gbps at the frequency range of 6GHz to 100GHz. Its application of Multiple Input, Multiple Output (MIMO) antenna technology enables it to beam dedicated radio signals to 5G enabled User Equipment (UEs) using the advanced beam steering algorithm. This results in maximum data throughput with ultra-high bandwidth transmission to end-user applications resulting in new and never-seen-before business opportunities across different industries, investments, and business services. By expanding the scope of wireless technologies and making devices more autonomous, 5G has the potential to reduce carbon emission footprint and encourage greener communities. The application of 5G cuts across all spheres of our lives; industry, transportation, education, communications, health, and businesses. 5G is more inclusive, progressive, proven, and powerful than any of its predecessor generations of communications technology. With more power and improved efficiency, 5G is on its way to creating a truly connected world.
Telekomünikasyon sektörü için veri madenciliği ve makine öğrenmesi teknikleri ile ayrılan müşteri analizi
(Düzce Üniversitesi, 2021) Uyanık, Furkan; Kasapbaşı, Mustafa Cem
Son yıllarda şirketler arası rekabetin artmasıyla beraber aboneliğinden ayrılacak müşterilerin tahmin edilmesi oldukça önemli hale gelmiştir. Müşteri karmaşası analizi, veri madenciliği, makine öğrenmesi ve derin öğrenme gibi alanlarda sıklıkla karşılaşılan analiz çeşitlerinden biridir. Özellikle telekomünikasyon, sigortacılık ve bankacılık gibi sektörlerde yaygın olarak kullanılmaktadır. Bu çalışma da veri madenciliği ve makine öğrenmesi teknikleri ile aboneliğini sonlandırma ihtimali olan müşterileri tahmin etmeyi amaçlamaktadır. Çalışma Lojistik Regresyon (Logistic Regression), Karar Ağacı (Decision Tree), Yapay Sinir Ağları (Artificial Neural Network), Torbalama (Bagging) ve Artırma (Boosting) sınıflandırma modelleri kullanılarak arasından en iyi sonucu bulmayı önermiştir. Veri setinde sınıf dengesizliği olduğu için SMOTE (Synthetic Minority Oversampling Technique) ve ADASYN (Adaptive Synthetic Sampling Method) tekniği ile örnekleme yapılmıştır. Çalışmada, 2 adet tahmin modeli önerilmiştir ve önerilen tahmin modelleri Veri Seti, Veri Ön İşleme, Veri Örnekleme, Değerlendirme olarak 4 farklı aşamadan oluşmaktadır. Veri Ön İşleme aşamasında, kullanılmayan ve önemsiz özniteliklerin veri setinden çıkartılması, normalizasyon, şifreleme (encoding) ve aşırı örnekleme gibi birçok yöntem kullanılmıştır. Performans ölçütü olarak Doğruluk Oranı (Accuracy Rate), Geri Çağırma (Recall), Hassasiyet (Precision) ve Özgünlük (Specificity), Dengelenmiş Doğruluk Oranı ve ROC Eğrisi Altındaki Alan (ROC-AUC) değeri kullanılmıştır. Performans ölçütlerine bakıldığında önerilen en iyi tahmin modeli ADASYN örnekleme yöntemi kullanılan model olmuştur. Sınıflandırma yöntemi olarak en iyi sonucu veren LightGBM (Light Gradient Boosting Machine) tekniği olmuştur. Önerilen modeller arasında Veri Ön İşleme ve Veri Örnekleme aşamalarında farklılıklar bulunmaktadır. Bu çalışmada önerilen tahmin modellerinin eğitim süresi, benzer çalışmalara göre daha iyi performans sağladığı tespit edilmiştir. Ayrıca bu çalışmada, sadece 58 öznitelik kullanarak 172 öznitelik kullanan benzer çalışmaların başardığına çok yakın sonuçlar elde edilmiştir.
Smart Cities and Data Analytics for Intelligent Transportation Systems: An Analytical Model for Scheduling Phases and Traffic Lights at Signalized Intersections
(MDPI, 2021) Güneş, Fatih; Bayraklı, Selim; Zaim, Abdül Halim
This paper is intended to improve the performance of signalized intersections, one of the most important systems of traffic control explained within the scope of smart transportation systems. These structures, which have the main role in ensuring the order and flow of traffic, are alternative systems depending on the different methods and techniques used. Determining the operation principles of these systems requires an extremely careful and planned study, considering their important role. Performance outputs obtained from the queue analyses made in previous studies formed the input of this study. The most important techniques are used in the effective control of intersections, such as signal timing: in particular, the use of effective green time and order of the transitions between phases. In this research, a traffic-sensitive signalized intersection control system was designed with the suggested methods against these two problems. The sample intersections were selected from three cities with the highest population density as the case study area. In the analysis of the performance of the connection arms of the selected intersections, flow intensity data were taken into consideration, as well as the arrival and service rates. Based on this, the outputs of the two proposed models regarding the use of phase transitions and effective green durations were compared with two other adaptive control systems and their positive results were shared. The results showed that signalized intersections, operating with a well-planned and correctly chosen technique, better regulate density and queuing.
A design of an integrated cloud-based intrusion detection system with third party cloud service
(De Gruyter, 2021) Elmasry, Wisam; Akbulut, Akhan; Zaim, Abdül Halim
Although cloud computing is considered the most widespread technology nowadays, it still suffers from many challenges, especially related to its security. Due to the open and distributed nature of the cloud environment, this makes the cloud itself vulnerable to various attacks. In this paper, the design of a novel integrated Cloud-based Intrusion Detection System (CIDS) is proposed to immunise the cloud against any possible attacks. The proposed CIDS consists of five main modules to do the following actions: monitoring the network, capturing the traffic flows, extracting features, analyzing the flows, detecting intrusions, taking a reaction, and logging all activities. Furthermore an enhanced bagging ensemble system of three deep learning models is utilized to predict intrusions effectively. Moreover, a third-party Cloud-based Intrusion Detection System Service (CIDSS) is also exploited to control the proposed CIDS and provide the reporting service. Finally, it has been shown that the proposed approach overcomes all problems associated with attacks on the cloud raised in the literature.
Artificial intelligence–based COVID-19 detection using cough records
(İstanbul Üniversitesi Cerrahpaşa, 2021) Gökcen, Alpaslan; Karadağ, Bulut; Riva, Cengiz; Boyacı, Ali
In 2019, with the emergence of coronavirus disease 2019 (COVID-19) and its spread all over the world, many people were directly affected by the pandemic. As its spread increases, it is difficult to diagnose who is actually infected. In addition to continuing vaccination studies, some technological solutions are being used to try to control the virus. One of these technological solutions is presented in this study. The disease is detected using cough data through artificial intelligence (AI). To do this, an open source data set was used from the opensigma.mit.edu website. More than 20,000 cough records representing age, gender, geographic location, and COVID-19 status are available from this site. The AI model trained on cough detection achieved 79% COVID-19 accuracy with an F1 of 80%. With the designed AI-based mobile application, COVID-19 can be detected and monitored.
Benchmark effect of web search engines on text mining
(Murat GÖK, 2021) Toprak, Ahmet; Turan, Metin
There have been many studies about creating a dictionary and these studies have come from past to present with different methods and different analyzes. Especially with the emergence of the World Wide Web, efforts to create dictionary based on instant data have gained importance. Therefore, the performance of the web search engines directly effects the model which is using web documents for automatic dictionary creation. The web search engines were evaluated in terms of their suggested documents relationality to the query in the research. For this purpose, an automatic dictionary creating model using web documents were developed. First of all, the topic seed words are determined by the documents presented to the system initially. Search is executed by these seed words initially. Then TF-IDF metric was used as meaningful word selection method for returned first document. The top n meaningful words were selected from the highest TF-IDF values. The value of n was determined experimentally. When searching the web with these words added to the dictionary, new documents were suggesting by the web search engine. By repeating the process, experimental dictionaries of a certain size were obtained. By the way, the documents suggested by each web engine are generally different, so that the dictionary similarity produced from the top suggested documents can measure web engines performance of selecting relational documents. Hash similarity was used to evaluate dictionary performance. According to the results, dictionary with the 73.9% highest similarity for Google search engine, dictionary with the 68.7% highest similarity for Bing search engine and dictionary with the 60.5% highest similarity for Yandex search engine were produced.
Dağıtık mesajlaşma altyapısı kullanılarak büyük boyutlu verilerin gerçek zamanlı olarak işlenmesi
(Murat GÖK, 2021) Toprak, Ahmet; Zaim, Abdül Halim
Günümüzde IoT (Internet of Things – Nesnelerin İnterneti) cihazların kullanımındaki artış beraberinde yüksek yoğunluklu ve farklı çeşitte verilerin oluşmasına sebep olmuştur. IoT cihazlarından toplanan bu verilerin formatları, şekilleri ve yoğunlukları birbirinden tamamen farklıdır. Bu verilerin anlık olarak işlenmesi ve ilgili kullanıcıya anlık olarak iletilmesi gerekmektedir. Bu makalede, IoT cihazlarından elde edilen verilerin işlenmesi ve son kullanıcıya anlık olarak iletilmesi amacıyla bir model tasarlanmıştır. Çalışmada öncelikli olarak IoT cihazlarından toplanan yapısal olmayan veriler veri ön işleme adımlarına tabi tutulmuştur. Veri ön işleme adımları sonrası elde edilen verilerden anlamlı kelimeler tespit edilmiştir. Bu amaçla TF-IDF (Term Frequency?-Inverse Document Frequency) metrikleri uygulanmıştır. Anlamlı kelime tespiti sonrası her anlamlı kelime konusuna göre verileri anlık işlemek amacıyla RabbitMQ dağıtık mesaj işleme kuyruğuna yönlendirilmiştir. Böylece mesajların iletilmesi garanti altına alınmıştır. RabbitMQ kuyruğuna iletilen mesajların anlık olarak alınması ve işlenmesi amacıyla Apache Storm topolojisi kullanılmıştır. Garantili mesaj işleme alt yapısı kullanan Apache Storm topolojisi, mesajları RabbitMQ dağıtık kuyruklama teknolojisi üzerinden okuyup, yapması gereken işlem ve hesaplamaları yaptıktan sonra çıktıları Elasticsearch içerisinde saklamıştır. Apache Storm topolojisi içerisinde üretilen sonuçlar daha sonra REST (Representational State Transfer) mimarisi kullanılarak son kullanıcı ile paylaşılmıştır.

Güncel Gönderiler