1,355,119 research outputs found
Use of a convolutional neural network for the classification of microbeads in urban wastewater
No full textScientists are on the lookout for a practical model that can serve as a standard for sorting out, identifying, and characterizing microplastics which are common occurrences in water sources and wastewaters. The microbeads (MBs) used in cosmetics and discharged into the sewer systems after use cause substantial microplastics pollution in the receiving waters. Today, the use of plastic microbeads in cosmetics is banned. The existing use cases are to be discontinued within a few years. Yet, there are no restrictions regarding the use of microbeads in a number of industries, cleaning products, pharmaceuticals and medical practices. In this context, the determination and classification of MBs which had so far been discharged to water sources and which continue to be discharged, represent crucial problems. In this work, we examined a new approach for the classification of MBs based on microscopic images. For classification purposes, Convolutional Neural Network (CNN) -a Deep Learning algorithm- was employed, whereas GoogLeNet architecture served as the model. The network is built from scratch, and trained then after tested on a total of 42928 images containing MBs in 5 distinct cleansers. The study performed with the CNN which achieved a classification performance of 89% for MBs in wastewater. (C) 2018 Elsevier Ltd. All rights reserved
alpyurtsever/SKETCH: SKETCH-v1.1
No full textThis release implements the methods described in the following references. See references for more details.
[TYUC2017P] J.A. Tropp, A. Yurtsever, M. Udell and V. Cevher, Practical Sketching Algorithms for Low-Rank Matrix Approximation, Accepted to SIAM J. Matrix Anal. Appl., August 2017.
[TYUC2017Nys] J.A. Tropp, A. Yurtsever, M. Udell and V. Cevher. Fixed-Rank Approximation of a Positive-Semidefinite Matrix from Streaming Data. In Proc. 31st Conference on Neural Information Processing Systems (NIPS), Long Beach, CA, USA, December 2017.
[TYUC2019] J.A. Tropp, A. Yurtsever, M. Udell and V. Cevher. Streaming Low-Rank Matrix Approximation with an Application to Scientific Simulation
Tiny, shiny, and colorful microplastics: Are regular glitters a significant source of microplastics?
No full textUsed in significant volumes in make-up, craft activities, and -more recently- in textile products, glitters are among single-use plastics, and are often made of polyethylene terephthalate. Even though a wealth of studies focus on the sources of microplastics in the environment and biota, glitters produced in various countries, and used extensively in entertainment events, shows and carnivals around the globe, not to mention by virtually anyone in daily life settings, have been relatively ignored as a major source of microplastics. That is why the present study focuses specifically on plastic glitters, and attempts to track them in the environment, in a manner comparable to their use in forensic science where glitters are often used as trace evidence associating a suspect with a specific murder case. Doing so led to various pieces of evidence of the presence of glitters -arguably a stealthy source of microplastics - in samples taken from the environment at a wide range of locations around the world
A practical methodology for predicting electricity consumption of urban residential buildings
No full textIt is a challenging task for determining the emissions and energy-saving options for counties and countries. In order to overcome this issue, two new concepts are introduced in this study. These are residential base electricity consumption and residential base natural gas consumption. A new model that estimates the total energy consumption of urban residential buildings from the available electricity consumption data is developed. The model employs heating degree days (HDD) and cooling degree days (CDD) parameters to predict the energy consumption for heating and cooling purposes. A month with zero or almost zero values of both HDD and CDD is considered as \"base month.\" These months when there is neither heating nor cooling, plays a key role in this study. The proposed model is applied to a county with a population of over 62,000 located in Southern Turkey, and the total electricity consumption is calculated with 1.24% accuracy. The validity and accuracy of the model are tested by theoretical electricity calculations, coal, and wood consumption. Accordingly, while annual electrical heating is equal to 2618 kWh/year for an average house of 96 m(2), for the coal and wood burning households, the annual energy consumption values are 2190 and 2616 kWh/year, respectively
alpyurtsever/SketchyCGAL: SketchyCGAL-v1.0
No full textThis release includes the source code for the numerical experiments in our paper.
Yurtsever, A., Tropp, J. A., Fercoq, O., Udell, M., Cevher, V.,
Scalable Semidefinite Programming,
SIAM Journal on Mathematics of Data Science, 3(1):171-200, 202
alpyurtsever/NonconvexTOS: NonconvexTOS
No full textThis release includes the source code for the experiments in our paper:
Yurtsever, A., Mangalick, V., Sra, S., Three Operator Splitting with a Nonconvex Loss Function, International Conference on Machine Learning, 202
Neutral and ionic dopants in helium clusters: interaction forces for the Li-2(a(3)Sigma(+)(u))-He and Li-2(+)(X-2 Sigma(+)(g))-He complexes
No full textThe potential energy surface (PES) describing the interactions between Li-2((1)Sigma(u)(+)) and He-4 and an extensive study of the energies and str uctures of a set of small clusters, Li-2(He)(n), were reported in a previous series of publications. In the present work, the same analysis is extended to the case of the excited Li-2(a(3)Sigma(u)(+)) and of the ionized Li-2(+) (X(2)Sigma(g)(+)) moiety. Calculated interaction potentials for the two title syst ems and the corresponding fitting of the computed points are shown. For both surfaces the MP4 method with cc-pV5Z basis sets has been used to generate an e xtensive range of radial/angular coordinates of the two-dimensional PESs which describe rigid rotor molecular dopants interacting with one He partner
Glitters as a Source of Primary Microplastics: An Approach to Environmental Responsibility and Ethics
No full textThis paper is about glitters, one of the sources of primary microplastics, which, in turn, are deemed an emerging source of pollutants affecting the environment. Today, most glitters available on the market are essentially microplastics, as they are made of polyesters and are of a size smaller than 5mm. The tiny, shiny, decorative and colorful glitters are used in a wide range of products including but not limited to make-up or craft materials, clothing, shoes, bags, ornaments, and various objects. The marketing of micron-sized plastic materials, the environmental risks of which are no longer disputable, as disposable products clearly merits attention. People without substantial knowledge or awareness about microplastics pollution and sources thereof have been consuming these glitter products without giving it a second thought. Given their tasteless, odourless, invisible, durable, and last but certainly not least, ubiquitous characteristics, small microplastics nowadays came to pose a substantial threat for the environment and the biota, as a sneaky and persistent contaminant. Microplastic pollution is a completely anthropogenic one. In this context, the consumers should assume a careful and sensible attitude, and question all their consumption habits. On the other hand, the responsibilities of the manufacturers and regulators are also substantial. For, in the lack of any regulation or sanctions in the context of laws and legislation concerning a global problem, and in a context of scarce awareness on part of the society, the people would continue to buy and consume plastic products like glitters without really thinking about them, for as long as they are supplied to the market
alpyurtsever/SHCGM: SHCGM
No full text<p>This release includes the source code for the numerical experiments in our paper:
[LYFC2019] F. Locatello, A. Yurtsever, O. Fercoq, V. Cevher,
"Stochastic Conditional Gradient Method for Composite Convex Minimization"
Advances in Neural Information Processing Systems 32 (NeurIPS 2019).</p>
A Late Old Syrian Stela from Temple M at Tilmen Höyük
No full textEditio princeps della stele del dio della tempesta risalente al 1700 a.C., rinvenuta negli scavi dell'Alma Mater Studiorum a Tilmen
- …
