scholarly journals Deep learning for pollen allergy surveillance from twitter in Australia

2019 ◽  
Vol 19 (1) ◽  
Author(s):  
Jia Rong ◽  
Sandra Michalska ◽  
Sudha Subramani ◽  
Jiahua Du ◽  
Hua Wang
Keyword(s):  
Deep Learning ◽  
Health Informatics ◽  
Pollen Allergy ◽  
Cost Effective ◽  
Machine Learning Techniques ◽  
Domain Specific ◽  
Public Health Monitoring ◽  
Real World Problem ◽  
Wide Range ◽  
Media Platform

Abstract Background The paper introduces a deep learning-based approach for real-time detection and insights generation about one of the most prevalent chronic conditions in Australia - Pollen allergy. The popular social media platform is used for data collection as cost-effective and unobtrusive alternative for public health monitoring to complement the traditional survey-based approaches. Methods The data was extracted from Twitter based on pre-defined keywords (i.e. ’hayfever’ OR ’hay fever’) throughout the period of 6 months, covering the high pollen season in Australia. The following deep learning architectures were adopted in the experiments: CNN, RNN, LSTM and GRU. Both default (GloVe) and domain-specific (HF) word embeddings were used in training the classifiers. Standard evaluation metrics (i.e. Accuracy, Precision and Recall) were calculated for the results validation. Finally, visual correlation with weather variables was performed. Results The neural networks-based approach was able to correctly identify the implicit mentions of the symptoms and treatments, even unseen previously (accuracy up to 87.9% for GRU with GloVe embeddings of 300 dimensions). Conclusions The system addresses the shortcomings of the conventional machine learning techniques with manual feature-engineering that prove limiting when exposed to a wide range of non-standard expressions relating to medical concepts. The case-study presented demonstrates an application of ’black-box’ approach to the real-world problem, along with its internal workings demonstration towards more transparent, interpretable and reproducible decision-making in health informatics domain.

scholarly journals Developing an Open-Source Lightweight Game Engine with DNN Support

Electronics ◽  
2020 ◽  
Vol 9 (9) ◽  
pp. 1421
Author(s):  
Haechan Park ◽  
Nakhoon Baek
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Open Source ◽  
Programming Languages ◽  
Cost Effective ◽  
Machine Learning Techniques ◽  
Learning Technology ◽  
Game Engine ◽  
Learning Techniques ◽  
Technical Issues

With the growth of artificial intelligence and deep learning technology, we have many active research works to apply the related techniques in various fields. To test and apply the latest machine learning techniques in gaming, it will be very useful to have a light-weight game engine for quick prototyping. Our game engine is implemented in a cost-effective way, in comparison to well-known commercial proprietary game engines, by utilizing open source products. Due to its simple internal architecture, our game engine is especially beneficial for modifying and reviewing the new functions through quick and repetitive tests. In addition, the game engine has a DNN (deep neural network) module, with which the proposed game engine can apply deep learning techniques to the game features, through applying deep learning algorithms in real-time. Our DNN module uses a simple C++ function interface, rather than additional programming languages and/or scripts. This simplicity enables us to apply machine learning techniques more efficiently and casually to the game applications. We also found some technical issues during our development with open sources. These issues mostly occurred while integrating various open source products into a single game engine. We present details of these technical issues and our solutions.

scholarly journals Genesis of MEMS Accelerometers for Select the Optimal Accelerometer for Bio Applications

2019 ◽  
Vol 9 (2) ◽  
pp. 4962-4970
Keyword(s):  
Structural Design ◽  
Cost Effective ◽  
Consumer Electronics ◽  
Measurement Unit ◽  
Micro Electro Mechanical Systems ◽  
Domain Specific ◽  
Wide Range ◽  
Mems Accelerometers ◽  
Unexplored Area ◽  
Selection For

MEMS(Micro Electro Mechanical Systems) Accelerometer is most popular and successful MEMS device for its versatile and multipurpose applications in different fields capable of measuring wide range of parameters ranging from vibration, acceleration, tilt and shock1 . Accelerometers, individually have their own operation and are also used in conjunction with other devices like gyroscopes, switches to make unique special devices like Inertial Measurement unit, widely used in consumer electronics, giving MEMS Accelerometers the giant share in the whole MEMS market. They have arrived as successful marketed devices in various designs and served many application in the fields of automobile , navigation, bio medical and space applications2 . These devices have diversified structural design and work on different principles, making them worth a study in terms of classification, selection for suitable application. Having the advantage of being cost effective and of little form factor, these devices have greater scope in various domains where the mechanical movements are needed to be analyzed. There is a huge unexplored area of BIO MEMS into which Accelerometers have to stretch their boundaries making Human body specific applications akin and par the present research. The present paper consolidates the existing accelerometers focusing on their applications while analyzing them in the light of their significance in each field of application. There is a great space for domain specific designing of the accelerometers trending into the usage in the current times.

scholarly journals A cost-effective maize ear phenotyping platform enables rapid categorization and quantification of kernels

2020 ◽  
Author(s):  
Cedar Warman ◽  
Christopher M. Sullivan ◽  
Justin Preece ◽  
Michaela E. Buchanan ◽  
Zuzana Vejlupkova ◽  
...  
Keyword(s):  
Computer Vision ◽  
Deep Learning ◽  
Low Cost ◽  
Digital Camera ◽  
Cost Effective ◽  
Detection Model ◽  
Biological Phenomena ◽  
Effective Combination ◽  
Wide Range ◽  
Maize Ear

AbstractHigh-throughput phenotyping systems are powerful, dramatically changing our ability to document, measure, and detect biological phenomena. Here, we describe a cost-effective combination of a custom-built imaging platform and deep-learning-based computer vision pipeline. A minimal version of the maize ear scanner was built with low-cost and readily available parts. The scanner rotates a maize ear while a cellphone or digital camera captures a video of the surface of the ear. Videos are then digitally flattened into two-dimensional ear projections. Segregating GFP and anthocyanin kernel phenotype are clearly distinguishable in ear projections, and can be manually annotated using image analysis software. Increased throughput was attained by designing and implementing an automated kernel counting system using transfer learning and a deep learning object detection model. The computer vision model was able to rapidly assess over 390,000 kernels, identifying male-specific transmission defects across a wide range of GFP-marked mutant alleles. This includes a previously undescribed defect putatively associated with mutation of Zm00001d002824, a gene predicted to encode a vacuolar processing enzyme (VPE). We show that by using this system, the quantification of transmission data and other ear phenotypes can be accelerated and scaled to generate large datasets for robust analyses.One sentence summaryA maize ear phenotyping system built from commonly available parts creates images of the surface of ears and identifies kernel phenotypes with a deep-learning-based computer vision pipeline.

scholarly journals Multi-omics and deep learning provide a multifaceted view of cancer

2021 ◽  
Author(s):  
Bora Uyar ◽  
Jonathan Ronen ◽  
Vedran Franke ◽  
Gaetano Gargiulo ◽  
Altuna Akalin
Keyword(s):  
Deep Learning ◽  
Complex Disease ◽  
Dimensional Space ◽  
Machine Learning Techniques ◽  
Response To Treatment ◽  
Molecular Features ◽  
Learning Techniques ◽  
Wide Range ◽  
Efficient Machine ◽  
Cancer Types

Cancer is a complex disease with a large financial and healthcare burden on society. One hallmark of the disease is the uncontrolled growth and proliferation of malignant cells. Unlike Mendelian diseases which may be explained by a few genomic loci, a deeper molecular and mechanistic understanding of the development of cancer is needed. Such an endeavor requires the integration of tens of thousands of molecular features across multiple layers of information encoded in the cells. In practical terms, this implies integration of multi omics information from the genome, transcriptome, epigenome, proteome, metabolome, and even micro-environmental factors such as the microbiome. Finding mechanistic insights and biomarkers in such a high dimensional space is a challenging task. Therefore, efficient machine learning techniques are needed to reduce the dimensionality of the data while simultaneously discovering complex but meaningful biomarkers. These markers then can lead to testable hypotheses in research and clinical applications. In this study, we applied advanced deep learning methods to uncover multi-omic fingerprints that are associated with a wide range of clinical and molecular features of tumor samples. Using these fingerprints, we can accurately classify different cancer types, and their subtypes. Non-linear multi-omic fingerprints can uncover clinical features associated with patient survival and response to treatment, ranging from chemotherapy to immunotherapy. In addition, multi-omic fingerprints may be deconvoluted into a meaningful subset of genes and genomic alterations to support clinically relevant decisions.

scholarly journals An Effective Evaluation on Fault Detection in Solar Panels

Energies ◽  
2021 ◽  
Vol 14 (22) ◽  
pp. 7770
Author(s):  
Joshuva Arockia Dhanraj ◽  
Ali Mostafaeipour ◽  
Karthikeyan Velmurugan ◽  
Kuaanan Techato ◽  
Prem Kumar Chaurasiya ◽  
...  
Keyword(s):  
Power Generation ◽  
Alternative Energy ◽  
Solar Power ◽  
Cost Effective ◽  
Renewable Resource ◽  
Machine Learning Techniques ◽  
Future Research ◽  
Solar Power Generation ◽  
Energy Demands ◽  
Wide Range

The world’s energy consumption is outpacing supply due to population growth and technological advancements. For future energy demands, it is critical to progress toward a dependable, cost-effective, and sustainable renewable energy source. Solar energy, along with all other alternative energy sources, is a potential renewable resource to manage these enduring challenges in the energy crisis. Solar power generation is expanding globally as a result of growing energy demands and depleting fossil fuel reserves, which are presently the primary sources of power generation. In the realm of solar power generation, photovoltaic (PV) panels are used to convert solar radiation into energy. They are subjected to the constantly changing state of the environment, resulting in a wide range of defects. These defects should be discovered and remedied as soon as possible so that PV panels efficiency, endurance, and durability are not compromised. This paper focuses on five aspects, namely, (i) the various possible faults that occur in PV panels, (ii) the online/remote supervision of PV panels, (iii) the role of machine learning techniques in the fault diagnosis of PV panels, (iv) the various sensors used for different fault detections in PV panels, and (v) the benefits of fault identification in PV panels. Based on the investigated studies, recommendations for future research directions are suggested.

scholarly journals Deep Learning for Image Based Mango Leaf Disease Detection

2019 ◽  
Vol 8 (3S3) ◽  
pp. 54-56
Keyword(s):  
Deep Learning ◽  
Cost Effective ◽  
Control Measures ◽  
Disease Detection ◽  
New Techniques ◽  
Biennial Bearing ◽  
Detection Technology ◽  
Wide Range ◽  
And Control ◽  
Effective Case

Among world’s mango producing countries, India ranks first and account 50% of the world’s mango production. The mango fruit is popular because of its wide range of adaptability, high nutritional value, different variety, delicious taste and excellent flavor. The fruit contains vitamin A and vitamin C in a rich extent. The crop is prone to diseases like powdery mildew, anthracnose, die back, blight, red rust, sooty mould, etc. Disorders may also impact the plant in the absence of effective case and control measures. These include change of form, biennial bearing, fall of fruit, black top, clustering, etc. The farmer must consult and take professional support for the prevention / control of diseases and crop disorder. New techniques of detecting mango disease are required to promote better control to avoid this crisis. By considering this, paper describes image recognition which provides cost effective and scalable disease detection technology. Paper further describes new deep learning models which give an opportunity for easy deployment of this technology. By considering a dataset of mango disease, pictures are taken from Konkan area in India. Transfer learning technique is used to train a profound Convolutionary Neural Network (CNN) to recognize 91% accuracy.

scholarly journals p1-FP: Extraction, Classification, and Prediction of Website Fingerprints with Deep Learning

2019 ◽  
Vol 2019 (3) ◽  
pp. 191-209 ◽  
Author(s):  
Se Eun Oh ◽  
Saikrishna Sunkam ◽  
Nicholas Hopper
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Deep Neural Network ◽  
State Of The Art ◽  
Machine Learning Techniques ◽  
Feature Engineering ◽  
Engineering Process ◽  
Learning Techniques ◽  
Wide Range

Abstract Recent advances in Deep Neural Network (DNN) architectures have received a great deal of attention due to their ability to outperform state-of-the-art machine learning techniques across a wide range of application, as well as automating the feature engineering process. In this paper, we broadly study the applicability of deep learning to website fingerprinting. First, we show that unsupervised DNNs can generate lowdimensional informative features that improve the performance of state-of-the-art website fingerprinting attacks. Second, when used as classifiers, we show that they can exceed performance of existing attacks across a range of application scenarios, including fingerprinting Tor website traces, fingerprinting search engine queries over Tor, defeating fingerprinting defenses, and fingerprinting TLS-encrypted websites. Finally, we investigate which site-level features of a website influence its fingerprintability by DNNs.

scholarly journals Emotional quantification of soundscapes by learning between samples

2020 ◽  
Vol 79 (41-42) ◽  
pp. 30387-30395
Author(s):  
Stavros Ntalampiras
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Deep Learning ◽  
State Of The Art ◽  
Emotional Responses ◽  
Machine Learning Techniques ◽  
Specific Domain ◽  
Learning Framework ◽  
Learning Techniques ◽  
Wide Range

Abstract Predicting the emotional responses of humans to soundscapes is a relatively recent field of research coming with a wide range of promising applications. This work presents the design of two convolutional neural networks, namely ArNet and ValNet, each one responsible for quantifying arousal and valence evoked by soundscapes. We build on the knowledge acquired from the application of traditional machine learning techniques on the specific domain, and design a suitable deep learning framework. Moreover, we propose the usage of artificially created mixed soundscapes, the distributions of which are located between the ones of the available samples, a process that increases the variance of the dataset leading to significantly better performance. The reported results outperform the state of the art on a soundscape dataset following Schafer’s standardized categorization considering both sound’s identity and the respective listening context.

2D Color-Intensity Representation of Ultrasonic Thickness Gauge Measurements

2020 ◽  
pp. 1192-1198
Author(s):  
M.S. Mohammad ◽  
Tibebe Tesfaye ◽  
Kim Ki-Seong
Keyword(s):  
Cost Effective ◽  
The Other ◽  
Thickness Gauge ◽  
Color Intensity ◽  
Digital Readout ◽  
Flat Surfaces ◽  
Video Cameras ◽  
Wide Range ◽  
Cost Effective Alternative ◽  
Ultrasonic Thickness

Ultrasonic thickness gauges are easy to operate and reliable, and can be used to measure a wide range of thicknesses and inspect all engineering materials. Supplementing the simple ultrasonic thickness gauges that present results in either a digital readout or as an A-scan with systems that enable correlating the measured values to their positions on the inspected surface to produce a two-dimensional (2D) thickness representation can extend their benefits and provide a cost-effective alternative to expensive advanced C-scan machines. In previous work, the authors introduced a system for the positioning and mapping of the values measured by the ultrasonic thickness gauges and flaw detectors (Tesfaye et al. 2019). The system is an alternative to the systems that use mechanical scanners, encoders, and sophisticated UT machines. It used a camera to record the probe’s movement and a projected laser grid obtained by a laser pattern generator to locate the probe on the inspected surface. In this paper, a novel system is proposed to be applied to flat surfaces, in addition to overcoming the other limitations posed due to the use of the laser projection. The proposed system uses two video cameras, one to monitor the probe’s movement on the inspected surface and the other to capture the corresponding digital readout of the thickness gauge. The acquired images of the probe’s position and thickness gauge readout are processed to plot the measured data in a 2D color-coded map. The system is meant to be simpler and more effective than the previous development.

Efficient Prediction of Structural and Electronic Properties of Hybrid 2D Materials Using DFT and Machine Learning

2018 ◽  
Author(s):  
Sherif Tawfik ◽  
Olexandr Isayev ◽  
Catherine Stampfl ◽  
Joseph Shapter ◽  
David Winkler ◽  
...  
Keyword(s):  
Machine Learning ◽  
Band Gap ◽  
Density Functional ◽  
2D Materials ◽  
Van Der Waals ◽  
Building Blocks ◽  
Machine Learning Techniques ◽  
Interlayer Distance ◽  
Computational Screening ◽  
Wide Range

Materials constructed from different van der Waals two-dimensional (2D) heterostructures offer a wide range of benefits, but these systems have been little studied because of their experimental and computational complextiy, and because of the very large number of possible combinations of 2D building blocks. The simulation of the interface between two different 2D materials is computationally challenging due to the lattice mismatch problem, which sometimes necessitates the creation of very large simulation cells for performing density-functional theory (DFT) calculations. Here we use a combination of DFT, linear regression and machine learning techniques in order to rapidly determine the interlayer distance between two different 2D heterostructures that are stacked in a bilayer heterostructure, as well as the band gap of the bilayer. Our work provides an excellent proof of concept by quickly and accurately predicting a structural property (the interlayer distance) and an electronic property (the band gap) for a large number of hybrid 2D materials. This work paves the way for rapid computational screening of the vast parameter space of van der Waals heterostructures to identify new hybrid materials with useful and interesting properties.

Sign in / Sign up

Export Citation Format

Share Document