105 research outputs found
SSFs dataset: Self-supervised features dataset of the ultrasonic signals of two-phase flow in an S-shaped riser
No full textTitle: SSFs dataset: Self-supervised features dataset of the ultrasonic signals of two-phase flow in an S-shaped riser
Author: Boyu Kuang, [email protected]
Time: 09th March 2023
Description:
The source data of the proposed SSFs dataset comes from: https://doi.org/10.17862/cranfield.rd.11369379.v1
The Self-supervised features (SSFs) dataset is opened to the community along with our latest journal paper entitled:
"Self-supervised learning-based two-phase flow regimen identification using ultrasonic sensors in an S-shape riser".
NOTE: the journal DOI will be provided after the acceptance.
This dataset is produced using the settings in TABLE I.
Here are some details, and please contact me if you got any issues with using the dataset:
SSFs_dataset: "the root directory of the dataset"
|
|
-- ex: "the SSFs from the experiment group (ex)"
| |
| |
| -- train: "the training set (70%)"
| |
| |
| -- test: "the testing set (15%)"
| |
| |
| -- valid: "the validation set (the rest)"
|
|
-- ctr-A: "the SSFs from the control group (ctr-A)"
| |
| |
| -- train: "the training set (70%)"
| |
| |
| -- test: "the testing set (15%)"
| |
| |
| -- valid: "the validation set (the rest)"
|
|
-- ... ...</p
Author recognition for Turkish documents
No full textGünümüzde, yazar tanıma çalışmaları, teknolojinin gelişmesi ve bilginin yaygınlaşması ile ortaya çıkan bir takım sorunlara çözüm üretmek için yapılmaktadır. Bu sorunlardan bazıları yazarı belli olmayan dokümanların yazarlarının belirlenmesi ve yazarının kim olduğundan tam olarak emin olunamayan metinlerin yazarlarının belirlenmesidir. Bu çalışmada, Türkçe dokümanlar için yazar tanıma sistemleri geliştirilmiştir. Sistemlerin eğitilmesinde ve test edilmesinde kullanılmak üzere, gazetelerden seçilen 6 yazara ait köşe yazıları kullanılmıştır. Yazarların 70?er makalesinden oluşan 420 dokümandan oluşan bir derlem hazırlanmıştır. Bu dokümanlardan 20?şer tanesi eğitim için, 50?şer tanesi test için kullanılmıştır. İlk olarak, 6 yazara ait dokümanlar toplanmış, daha sonra her yazara ait 20 doküman birleştirilerek tek bir doküman haline getirilmiştir. Bu şekilde elde edilen 6 doküman için sözcük, gövde, hece ve karakter n-gramlarının öznitelik vektörleri belirlenmiştir. K-En Yakın Komşu algoritması için öznitelik vektörleri belirlenirken her yazar için vektör uzunlukları 120, 180 ve 240 olarak seçilmiş, oluşan öznitelik vektörleri için K-En Yakın Komşu algoritmasıyla test edilmiştir. En başarılı sonuçlar, vektör boyu 120 olduğunda elde edildiğinden diğer metotlar için de vektör boyu 120 olarak kullanılmıştır. Geliştirilen sistemler eğitildikten sonra test edilerek doğruluk ve F-ölçüsü değerlerine göre birbirleriyle karşılaştırılmıştır.Today, the studies of author recognition have been made for providing the solutions of the problems which occur by developing and growing of information technology. Some of these problems are to specify the authors who the papers are exactly written by. In this study, some systems about author recognition for Turkish documents have been developed. For generating the systems, we have used the columns which belong to six authors in some newspapers. A corpus which includes totally 420 documents is constructed for training and testing of the systems. Each author has seventy documents. Twenty documents of every author are used for training operation. But, the other documents are utilized for testing stage. The features of word, stem, syllable, character and their n-grams are decided for each documents of these six author. Author recognition systems have been developed with the methods as K-Nearest Neighbor, Support Vector Machine, Multi-Layer Perceptron and Learning Vector Quantization. The feature vectors? lengths of the systems developed by K-Nearest Neighbor have been chosen as 120, 180 and 240. Because the most successful results are obtained as the length of the feature vectors is 120, we have used this length for the other methods. After the developed systems are trained the methods, the systems have been tested and evaluated according to accuracy and F-measure values
Examining Changes in the Democratic Progressive Party from 2008 to 2014 with Harmel and Janda\ue2s Integrated Theory of Party Change
No full textUsing Harmel and Janda\ue2s integrated theory of party change, the author examines the changes within the Democratic Progressive Party\ue3(DPP)\ue3from 2008-2014. In this thesis, the author specifically explores how the three main independent variables \ue2 chairperson change, dominant faction displacement, and external stimuli \ue2 are related to the changes observed in DPP which is considered a vote-seeking party.
The research finds that changes within the DPP including organization, ideology, strategies, constitution, and policy changes can be attributed to numerous factors such as \ue2 changes in DDP\ue2s chairperson, external stimuli and external shocks such as electoral system change, jurisdictional changes in Taiwan\ue2s local government structure, the Sunflower protest movement in early 2014, and others. Despite these changes, DPP factional structure have not changed substantially. The changes observed in the DPP in this thesis corroborates Harmel and Janda\ue2s integrated theory of party change
A Novel Quercetin Encapsulated Glucose Modified Liposome and Its Brain-Target Antioxidative Neuroprotection Effects
No full textNeurodegenerative diseases (NDDs) are mainly induced by oxidative stress which produces excessive reactive oxygen species (ROS). Quercetin (QU) is a potent antioxidant with some effects on NDDs. This study prepared and characterized a novel glucose-modified QU liposome (QU–Glu–Lip), aiming not only to overcome QU’s poor water solubility and bioavailability but also to deliver more QU to brain tissue to enhance its neuroprotective effect. QU–Glu–Lip possessed encapsulation efficiency (EE) of 89.9%, homogenous particle sizes (116–124 nm), small PDI value (<0.3), zeta value −1.363 ± 0.437 mV, proper pH and salt stability, and proper cytotoxicity. The glucose-modified liposome penetrated the blood–brain barrier (BBB) mediated via the glucose transporter 1 (GLUT1) and was taken by neuronal cells more efficiently than liposome without glucose, according to bEnd.3 and PC12 cell tests. QU–Glu–Lip attenuated H2O2-induced oxidative damage to PC12 with higher cell viability (88.42%) and lower intracellular ROS compared to that of QU. QU–Glu–Lip had higher brain target ability and delivered more QU to neuronal cells, effectively exerting the antioxidative neuroprotection effect. There is potential for the QU–Glu–Lip application for more effective treatment of NDDs
Design of optoacoustic imaging system for care medical diagnostics
No full textThis report presents what the author have done for the Final Year Project. Firstly, involved the design of a palm size laser diode driver circuit board which later on could be used in the photoacoustic imaging system for driving a low power laser diode. Electrical circuit design, circuit simulation on LTSpice, PCB layout design and hands on experiments were also performed.
Secondly, studied the photoacoustic effect and furthermore a photoacoustic imaging system was implemented to achieve high spatial resolution and good contrast image.
Last but not least, had a basic understanding of photoacoustic effects, coupled with the research study on PA-SAW devices, a PA-SAW based particle sensing system was designed to effective enhance the sensing signal.Bachelor of Engineerin
Research on the Optimal Machine Learning Classifier for Traffic Signs
No full textNow autonomous driving is a hot topic, and the identification of traffic signs is also extremely important for autonomous driving. This paper mainly compares the difference of the Support Vector Machine (SVM), Multilayer Perceptron (MLP), and Logistic Regression (LR) Classifier in the traffic sign classification. The effect of the initial image processing on classification accuracy is also studied. The paper found that sharpening the image significantly improved the accuracy of the image classification. Based on the results of various situations, the author found that, in this paper, SVM is the classifier with the best classification effect, but the effect of LR classifier is not much worse than that of SVM when the image is sharpened
Webpage ranking : how it affects your selections
No full textWhat do you need to search for your desired answer? Recalling the period before 1997, the most possible would be “go the libraries”. And now, the most possible answer is a computer with internet connected to find it out on the search engines. Web site like Google, Bing and Yahoo etc. are all commonly used search engine. These search engines will provide long lists of website for each query. How would the user decide the result they really desire? According to the research results from Chitika, a worldwide search-targeted network, the highest Google result impressions percentage is the result listed in the first position at 34.5%. As shown in the result, webpage ranking greatly affect the choices made by the users.
In this project, the author will use an implicit search result data collection website which will collect the implicit information during the interviewee making choices after search queries from google.com. The collected data will give a better understanding of how webpage ranking affect the choices under different scenario.
The project was built using Apache Web Server, MySQL Database, PHP and HTML to create an intranet based implicit search result data collection website. SQLyogEnt was involved for a better management of the data collected.Bachelor of Engineerin
Mitigating the autogenous shrinkage of alkali-activated slag by metakaolin
No full textThis study investigates the effectiveness of metakaolin (MK)in mitigating the autogenous shrinkage of alkali-activated slag (AAS). It is found that the autogenous shrinkage of AAS paste can be reduced by 40% and 50% when replacing 10% and 20% slag with MK, respectively. By providing additional Si and Al, and decreasing the pH of the pore solution, the incorporation of MK retards the formation of aluminium-modified calcium silicate hydrate (CASH)gels, the main reaction products in the studied pastes. The chemical shrinkage and pore refinement are consequently mitigated, resulting in a substantial reduction in the pore pressure. Meanwhile, the elastic modulus of AAS paste is only slightly influenced after MK addition. As a result, the autogenous shrinkage of AAS is significantly mitigated by incorporating MK. In addition, the introduction of MK would extend the setting time, slightly decrease the compressive strength, but greatly increase the flexural strength of AAS.Accepted Author ManuscriptMaterials and Environmen
Deprem ve diyaliz : bibliyometrik analiz
No full textAim: This research was conducted with the aim of identifying and visualizing articles on the relationship between dialysis and earthquakes, highlighting trends, and providing future researchers with a literaturebased overview.
Methods: The data were obtained from the "Web of Science Core Collection" data base on November 20, 2023, using the keywords 'dialysis' and 'earthquake,' resulting in 94 research articles from a population of 138 studies. Bibliometric analysis was carried out using the "VOSviewer" and "R-based Bibliyometrix" programs, including performance analysis and scientific mapping.
Results: As a result of the analysis, it was determined that the most frequently used keywords in the Web of Science category were ‘acute-renal-failure,‘management’,‘dialysis’, ‘earthquake’, ‘acute kidney injury’, ‘crush syndrome’. When the countries of publication were evaluated, it was determined that the most studies were conducted in 2009 and in Türkiye, and the author who published the most and contributed the most was ‘Vanholder R.’. It was determined that the first three most frequently used keywords in the publications were ‘crush syndrome’, ‘dialysis’ and ‘acuterenal failure’.
Conclusion: The findings of this study are believed to contribute to evaluating the management of dialysis patients in the event of a possible earthquake and guiding future research planning.Amaç: Bu araştırma, diyaliz ve deprem arasındaki ilişkiye ilişkin makaleleri belirlemek ve görselleştirmek, eğilimleri vurgulamak ve gelecekteki araştırmacılara literatüre dayalı bir genel bakış sağlamak amacıyla yapılmıştır.
Yöntemler: Veriler, 20 Kasım 2023 tarihinde "Web of Science Core Collection" veritabanından 'diyaliz' ve 'deprem' anahtar kelimeleri kullanılarak elde edildi ve sonuçta 138 çalışmadan oluşan bir popülasyondan 94 araştırma makalesi elde edildi. Bibliyometrik analiz, performans analizi ve bilimsel haritalamayı içeren "VOSviewer" ve "Rtabanlı Bibliyometrix" programları kullanılarak gerçekleştirildi.
Bulgular: Yapılan analiz sonucunda Web of Science kategorisinde en çok kullanılan anahtar kelimenin “acute-renal-failure”, “management”, “dialysis”, "earthquake”, “acute kidney injury”, “crush sendrome”
olduğu belirlenmiştir. Yayın yapılan ülkeler değerlendirildiğinde en fazla çalışmanın 2009 yılında ve Türkiye'de yapıldığı, en fazla yayın yapan ve en fazla katkı sağlayan yazarın “Vanholder R." olduğu saptanmıştır. Yayınlarda en sık kullanılan ilk üç anahtar kelimenin “crush sendromu”, “diyaliz” ve
“akut-böbrek yetmezliği” olduğu belirlenmiştir.
Sonuç: Bu çalışmanın bulgularının olası bir deprem durumunda diyaliz hastalarının yönetiminin değerlendirilmesine ve gelecekteki araştırma planlamalarına yol göstermesine katkı sağlayacağı
düşünülmektedir
- …
