Although neural networks (NN) are important especially for engineers, scientists, mathematicians and statisticians, they may also be hard to understand. In this article, application areas of NN are discussed, basic NN components are described and it is explained how an NN work. A web-based simulation and visualization tool (EasyLearnNN) is developed using Java and Java 2D for teaching NN concepts. Perceptron, ADALINE, Multilayer Perceptron, LVQ and SOM models and related training algorithms are implemented. As a result, comparison with other teaching methods of NN concepts is presented and discussed. ß
In geospatial applications such as urban planning and land use management, automatic detection and classification of earth objects are essential and primary subjects. When the significant semantic segmentation algorithms are considered, DeepLabV3+ stands out as a state-of-the-art CNN. Although the DeepLabV3+ model is capable of extracting multi-scale contextual information, there is still a need for multi-stream architectural approaches and different training approaches of the model that can leverage multi-modal geographic datasets. In this study, a new end-to-end dual-stream architecture that considers geospatial imagery was developed based on the DeepLabV3+ architecture. As a result, the spectral datasets other than RGB provided increments in semantic segmentation accuracies when they were used as additional channels to height information. Furthermore, both the given data augmentation and Tversky loss function which is sensitive to imbalanced data accomplished better overall accuracies. Also, it has been shown that the new dual-stream architecture using Potsdam and Vaihingen datasets produced 88.87% and 87.39% overall semantic segmentation accuracies, respectively. Eventually, it was seen that enhancement of the traditional significant semantic segmentation networks has a great potential to provide higher model performances, whereas the contribution of geospatial data as the second stream to RGB to segmentation was explicitly shown.
Although the source active regions of some coronal mass ejections (CMEs) were identified in CME catalogues, vast majority of CMEs do not have an identified source active region. We propose a method that uses a filtration process and machine learning to identify the sunspot groups associated with a large fraction of CMEs and compare the physical parameters of these identified sunspot groups with properties of their corresponding CMEs to find mechanisms behind the initiation of CMEs. These CMEs were taken from the Coordinated Data Analysis Workshops (CDAW) database hosted at NASA's website. The Helio-seismic and Magnetic Imager (HMI) Active Region Patches (HARPs) were taken from the Stanford University's JSOC database. The source active regions of the CMEs were identified by the help of a custom filtration procedure and then by training a Long Short-Term Memory Network (LSTM) to identify the patterns in the physical magnetic parameters derived from vector and line of sight magnetograms. The neural network simultaneously considers the time series data of these magnetic parameters at once and learns the patterns at the onset of CMEs. This neural network was then used to identify the source HARPs for the CMEs recorded from 2011 till 2020. The neural network was able to reliably identify source HARPs for 4895 CMEs out of 14604 listed in the CDAW database during the afore-mentioned period.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.