Search results for: custom dataset

Authors: Merve Yünlü, Nihat Kandemir, Aylin Ersoy

Abstract:

Temperature sensors are used in home appliances not only to monitor the basic functions of the machine but also to minimize energy consumption and ensure safe operation. In parallel with the development of smart home applications and IoT algorithms, these sensors produce important data such as the frequency of use of the machine, user preferences, and the compilation of critical data in terms of diagnostic processes for fault detection throughout an appliance's operational lifespan. Commercially available thin-film resistive temperature sensors have a well-established manufacturing procedure that allows them to operate over a wide temperature range. However, these sensors are over-designed for white goods applications. The operating temperature range of these sensors is between -70°C and 850°C, while the temperature range requirement in home appliance applications is between 23°C and 500°C. To ensure the operation of commercial sensors in this wide temperature range, usually, a platinum coating of approximately 1-micron thickness is applied to the wafer. However, the use of platinum in coating and the high coating thickness extends the sensor production process time and therefore increases sensor costs. In this study, an attempt was made to develop a low-cost temperature sensor design and production method that meets the technical requirements of white goods applications. For this purpose, a custom design was made, and design parameters (length, width, trim points, and thin film deposition thickness) were optimized by using statistical methods to achieve the desired resistivity value. To develop thin film resistive temperature sensors, one side polished sapphire wafer was used. To enhance adhesion and insulation 100 nm silicon dioxide was coated by inductively coupled plasma chemical vapor deposition technique. The lithography process was performed by a direct laser writer. The lift-off process was performed after the e-beam evaporation of 10 nm titanium and 280 nm platinum layers. Standard four-point probe sheet resistance measurements were done at room temperature. The annealing process was performed. Resistivity measurements were done with a probe station before and after annealing at 600°C by using a rapid thermal processing machine. Temperature dependence between 25-300 °C was also tested. As a result of this study, a temperature sensor has been developed that has a lower coating thickness than commercial sensors but can produce reliable data in the white goods application temperature range. A relatively simplified but optimized production method has also been developed to produce this sensor.

Keywords: thin film resistive sensor, temperature sensor, household appliance, sustainability, energy efficiency

Procedia PDF Downloads 52

Authors: Dongju Kim, Youngjoo Suh, Hyojin Kim, Gyeongyeong Kim

Abstract:

Acoustic Emission Testing (AET) is widely used to test the structural integrity of an operational hydrogen storage container, and clustering algorithms are frequently used in pattern recognition methods to interpret AET results. However, the interpretation of AET results can vary from user to user as the tuning of the relevant parameters relies on the user's experience and knowledge of AET. Therefore, it is necessary to use a deep learning model to identify patterns in acoustic emission (AE) signal data that can be used to classify defects instead. In this paper, a deep learning-based model for classifying the types of defects in hydrogen storage tanks, using AE sensor waveforms, is proposed. As hydrogen storage tanks are commonly constructed using carbon fiber reinforced polymer composite (CFRP), a defect classification dataset is collected through a tensile test on a specimen of CFRP with an AE sensor attached. The performance of the classification model, using one-dimensional convolutional neural network (1-D CNN) and synthetic minority oversampling technique (SMOTE) data augmentation, achieved 91.09% accuracy for each defect. It is expected that the deep learning classification model in this paper, used with AET, will help in evaluating the operational safety of hydrogen storage containers.

Keywords: acoustic emission testing, carbon fiber reinforced polymer composite, one-dimensional convolutional neural network, smote data augmentation

Procedia PDF Downloads 70

Authors: Ons Slimene, Aroua Taamallah, Maha Khemaja

Abstract:

In recent years, human activity recognition (HAR) has been a key area of research due to its diverse applications. It has garnered increasing attention in the field of computer vision. HAR plays an important role in people’s daily lives as it has the ability to learn advanced knowledge about human activities from data. In HAR, activities are usually represented by exploiting different types of sensors, such as embedded sensors or visual sensors. However, these sensors have limitations, such as local obstacles, image-related obstacles, sensor unreliability, and consumer concerns. Recently, several deep learning-based approaches have been proposed for HAR and these approaches are classified into two categories based on the type of data used: vision-based approaches and sensor-based approaches. This research paper highlights the importance of multimodal data fusion from skeleton data obtained from videos and data generated by embedded sensors using deep neural networks for achieving HAR. We propose a deep multimodal fusion network based on a twostream architecture. These two streams use the Convolutional Neural Network combined with the Bidirectional LSTM (CNN BILSTM) to process skeleton data and data generated by embedded sensors and the fusion at the feature level is considered. The proposed model was evaluated on a public OPPORTUNITY++ dataset and produced a accuracy of 96.77%.

Keywords: human activity recognition, action recognition, sensors, vision, human-centric sensing, deep learning, context-awareness

Procedia PDF Downloads 76

Authors: Sushish Baral, Pratibha Joshi, Yaman Maharjan

Abstract:

Glaucoma is a chronic eye condition that causes vision loss that is irreversible. Early detection and treatment are critical to prevent vision loss because it can be asymptomatic. For the identification of glaucoma, multiple deep learning algorithms are used. Transformer-based architectures, which use the self-attention mechanism to encode long-range dependencies and acquire extremely expressive representations, have recently become popular. Convolutional architectures, on the other hand, lack knowledge of long-range dependencies in the image due to their intrinsic inductive biases. The aforementioned statements inspire this thesis to look at transformer-based solutions and investigate the viability of adopting transformer-based network designs for glaucoma detection. Using retinal fundus images of the optic nerve head to develop a viable algorithm to assess the severity of glaucoma necessitates a large number of well-curated images. Initially, data is generated by augmenting ocular pictures. After that, the ocular images are pre-processed to make them ready for further processing. The system is trained using pre-processed images, and it classifies the input images as normal or glaucoma based on the features retrieved during training. The Vision Transformer (ViT) architecture is well suited to this situation, as it allows the self-attention mechanism to utilise structural modeling. Extensive experiments are run on the common dataset, and the results are thoroughly validated and visualized.

Keywords: glaucoma, vision transformer, convolutional architectures, retinal fundus images, self-attention, deep learning

Procedia PDF Downloads 170

Authors: Christoph Opperer

Abstract:

Space syntax is a theory and method for analyzing the spatial layout of buildings and urban environments to understand how they can influence patterns of human movement, social interaction, and behavior. While direct visibility is a key factor in space syntax research, important visual information such as light, color, texture, etc., are typically not considered, even though psychological studies have shown a strong correlation to the human perceptual experience within physical space – with light and color, for example, playing a crucial role in shaping the perception of spaciousness. Furthermore, these surface properties are often the visual features that are most salient and responsible for drawing attention to certain elements within the environment. This paper explores the potential of integrating these factors into general space syntax methods and visibility-based analysis of space, particularly for architectural spatial layouts. To this end, we use a combination of geometric (isovist) and topological (visibility graph) approaches together with image-based methods, allowing a comprehensive exploration of the relationship between spatial geometry, visual aesthetics, and human experience. Custom-coded ray-tracing techniques are employed to generate spherical panorama images, encoding three-dimensional spatial data in the form of two-dimensional images. These images are then processed through computer vision algorithms to generate saliency-maps, which serve as a visual representation of areas most likely to attract human attention based on their visual properties. The maps are subsequently used to weight the vertices of isovists and the visibility graph, placing greater emphasis on areas with high saliency. Compared to traditional methods, our weighted visibility analysis introduces an additional layer of information density by assigning different weights or importance levels to various aspects within the field of view. This extends general space syntax measures to provide a more nuanced understanding of visibility patterns that better reflect the dynamics of human attention and perception. Furthermore, by drawing parallels to traditional isovist and VGA analysis, our weighted approach emphasizes a crucial distinction, which has been pointed out by Ervin and Steinitz: the difference between what is possible to see and what is likely to be seen. Therefore, this paper emphasizes the importance of including surface properties in visibility-based analysis to gain deeper insights into how people interact with their surroundings and to establish a stronger connection with human attention and perception.

Keywords: space syntax, visibility analysis, isovist, visibility graph, visual features, human perception, saliency detection, raytracing, spherical images

Procedia PDF Downloads 50

Authors: Onyedikachi Ulelu, Andrew P. Longstaff, Simon Fletcher, Simon Parkinson

Abstract:

The widespread interest in 'Industry 4.0' or 'digital manufacturing' has led to significant research requiring the acquisition of data from sensors, instruments, and machine signals. In-depth research then identifies methods of analysis of the massive amounts of data generated before and during manufacture to solve a particular problem. The ultimate goal is for industrial Internet of Things (IIoT) data to be processed automatically to assist with either visualisation or autonomous system decision-making. However, the collection and processing of data in an industrial environment come with a cost. Little research has been undertaken on how to specify optimally what data to capture, transmit, process, and store at various levels of an edge-cloud network. The first step in this specification is to categorise IIoT data for efficient and effective use. This paper proposes the required attributes and classification to take manufacturing digital data from various sources to determine the most suitable location for data processing on the edge-cloud network. The proposed classification framework will minimise overhead in terms of network bandwidth/cost and processing time of machine tool data via efficient decision making on which dataset should be processed at the ‘edge’ and what to send to a remote server (cloud). A fast-and-frugal heuristic method is implemented for this decision-making. The framework is tested using case studies from industrial machine tools for machine productivity and maintenance.

Keywords: data classification, decision making, edge computing, industrial IoT, industry 4.0

Procedia PDF Downloads 156

Authors: Youngjae Jin, Daeshik Kim

Abstract:

This paper describes a cycle accurate simulation results of weight values learned by an auto-encoder behavior model in terms of pre-route simulation. Given the results we visualized the first layer representations with natural images. Many common deep learning threads have focused on learning high-level abstraction of unlabeled raw data by unsupervised feature learning. However, in the process of handling such a huge amount of data, the learning method’s computation complexity and time limited advanced research. These limitations came from the fact these algorithms were computed by using only single core CPUs. For this reason, parallel-based hardware, FPGAs, was seen as a possible solution to overcome these limitations. We adopted and simulated the ready-made auto-encoder to design a behavior model in Verilog HDL before designing hardware. With the auto-encoder behavior model pre-route simulation, we obtained the cycle accurate results of the parameter of each hidden layer by using MODELSIM. The cycle accurate results are very important factor in designing a parallel-based digital hardware. Finally this paper shows an appropriate operation of behavior model based pre-route simulation. Moreover, we visualized learning latent representations of the first hidden layer with Kyoto natural image dataset.

Keywords: auto-encoder, behavior model simulation, digital hardware design, pre-route simulation, Unsupervised feature learning

Procedia PDF Downloads 423

Authors: Stelios A. Mitilineos, Nicolas-Alexander Tatlas, Georgia Korompili, Lampros Kokkalas, Stelios M. Potirakis

Abstract:

Obstructive Sleep Apnea Hypopnea Syndrome (OSAHS) is a widespread chronic disease that mostly remains undetected, mainly due to the fact that it is diagnosed via polysomnography which is a time and resource-intensive procedure. Screening the disease’s symptoms at home could be used as an alternative approach in order to alert individuals that potentially suffer from OSAHS without compromising their everyday routine. Since snoring is usually linked to OSAHS, developing a snore detector is appealing as an enabling technology for screening OSAHS at home using ubiquitous equipment like commodity microphones (included in, e.g., smartphones). In this context, this study developed a snore detection tool and herein present the approach and selection of specific sound features that discriminate snoring vs. environmental sounds, as well as the performance of the proposed tool. Furthermore, a Real-Time Snore Detector (RTSD) is built upon the snore detection tool and employed in whole-night sleep sound recordings resulting to a large dataset of snoring sound excerpts that are made freely available to the public. The RTSD may be used either as a stand-alone tool that offers insight to an individual’s sleep quality or as an independent component of OSAHS screening applications in future developments.

Keywords: obstructive sleep apnea hypopnea syndrome, apnea screening, snoring detection, machine learning, neural networks

Procedia PDF Downloads 188

Authors: Farshid Hajati, Federico Girosi, Shima Ghassempour

Abstract:

We present a Deep Belief Network (DBN) method for clustering health trajectories. Deep Belief Network (DBN) is a deep architecture that consists of a stack of Restricted Boltzmann Machines (RBM). In a deep architecture, each layer learns more complex features than the past layers. The proposed method depends on DBN in clustering without using back propagation learning algorithm. The proposed DBN has a better a performance compared to the deep neural network due the initialization of the connecting weights. We use Contrastive Divergence (CD) method for training the RBMs which increases the performance of the network. The performance of the proposed method is evaluated extensively on the Health and Retirement Study (HRS) database. The University of Michigan Health and Retirement Study (HRS) is a nationally representative longitudinal study that has surveyed more than 27,000 elderly and near-elderly Americans since its inception in 1992. Participants are interviewed every two years and they collect data on physical and mental health, insurance coverage, financial status, family support systems, labor market status, and retirement planning. The dataset is publicly available and we use the RAND HRS version L, which is easy to use and cleaned up version of the data. The size of sample data set is 268 and the length of the trajectories is equal to 10. The trajectories do not stop when the patient dies and represent 10 different interviews of live patients. Compared to the state-of-the-art benchmarks, the experimental results show the effectiveness and superiority of the proposed method in clustering health trajectories.

Keywords: health trajectory, clustering, deep learning, DBN

Procedia PDF Downloads 347

Authors: Neha Ahirwar

Abstract:

In the contemporary digital era, the rise of credit card fraud poses a significant threat to both financial institutions and consumers. As fraudulent activities become more sophisticated, there is an escalating demand for robust and effective fraud detection mechanisms. Advanced machine learning algorithms have become crucial tools in addressing this challenge. This paper conducts a thorough examination of the design and evaluation of a credit card fraud detection system, utilizing four prominent machine learning algorithms: random forest, logistic regression, decision tree, and XGBoost. The surge in digital transactions has opened avenues for fraudsters to exploit vulnerabilities within payment systems. Consequently, there is an urgent need for proactive and adaptable fraud detection systems. This study addresses this imperative by exploring the efficacy of machine learning algorithms in identifying fraudulent credit card transactions. The selection of random forest, logistic regression, decision tree, and XGBoost for scrutiny in this study is based on their documented effectiveness in diverse domains, particularly in credit card fraud detection. These algorithms are renowned for their capability to model intricate patterns and provide accurate predictions. Each algorithm is implemented and evaluated for its performance in a controlled environment, utilizing a diverse dataset comprising both genuine and fraudulent credit card transactions.

Keywords: efficient credit card fraud detection, random forest, logistic regression, XGBoost, decision tree

Procedia PDF Downloads 35

Authors: Jean-Francois Rajotte, Martin Sotir, Frank Gouineau

Abstract:

The installation and management of traffic monitoring devices can be costly from both a financial and resource point of view. It is therefore important to take advantage of in-place infrastructures to extract the most information. Here we show how low-quality urban road traffic images from cameras already available in many cities (such as Montreal, Vancouver, and Toronto) can be used to estimate traffic flow. To this end, we use a pre-trained neural network, developed for object detection, to count vehicles within images. We then compare the results with human annotations gathered through crowdsourcing campaigns. We use this comparison to assess performance and calibrate the neural network annotations. As a use case, we consider six months of continuous monitoring over hundreds of cameras installed in the city of Montreal. We compare the results with city-provided manual traffic counting performed in similar conditions at the same location. The good performance of our system allows us to consider applications which can monitor the traffic conditions in near real-time, making the counting usable for traffic-related services. Furthermore, the resulting annotations pave the way for building a historical vehicle counting dataset to be used for analysing the impact of road traffic on many city-related issues, such as urban planning, security, and pollution.

Keywords: traffic monitoring, deep learning, image annotation, vehicles, roads, artificial intelligence, real-time systems

Procedia PDF Downloads 175

Authors: S. Abera, T. Gidey, W. Terefe

Abstract:

Introduction: Testing for HIV is the key entry point to HIV prevention, treatment, and care and support services. Hence, predictive data mining techniques can greatly benefit to analyze and discover new patterns from huge datasets like that of EDHS 2011 data. Objectives: The objective of this study is to build a predictive modeling for HIV testing and identify determinants of HIV testing for adults with age above fourteen years using data mining techniques. Methods: Cross-Industry Standard Process for Data Mining (CRISP-DM) was used to predict the model for HIV testing and explore association rules between HIV testing and the selected attributes among adult Ethiopians. Decision tree, Naïve-Bayes, logistic regression and artificial neural networks of data mining techniques were used to build the predictive models. Results: The target dataset contained 30,625 study participants; of which 16, 515 (53.9%) were women. Nearly two-fifth; 17,719 (58%), have never been tested for HIV while the rest 12,906 (42%) had been tested. Ethiopians with higher wealth index, higher educational level, belonging 20 to 29 years old, having no stigmatizing attitude towards HIV positive person, urban residents, having HIV related knowledge, information about family planning on mass media and knowing a place where to get testing for HIV showed an increased patterns with respect to HIV testing. Conclusion and Recommendation: Public health interventions should consider the identified determinants to promote people to get testing for HIV.

Keywords: data mining, HIV, testing, ethiopia

Procedia PDF Downloads 470

Authors: Huawei Bai, Jianguo Yao, Fellow, IEEE

Abstract:

Atrial fibrillation (AF) is the most common form of heart arrhythmia and is closely associated with mortality and morbidity in heart failure, stroke, and coronary artery disease. The development of single spot optical sensors enables widespread photoplethysmography (PPG) screening, especially for AF, since it represents a more convenient and noninvasive approach. To our knowledge, most existing studies based on public and unbalanced datasets can barely handle the multiple noises sources in the real world and, also, lack interpretability. In this paper, we construct a large- scale PPG dataset using measurements collected from PPG wrist- watch devices worn by volunteers and propose an attention-based two-stream heterogeneous neural network (TSHNN). The first stream is a hybrid neural network consisting of a three-layer one-dimensional convolutional neural network (1D-CNN) and two-layer attention- based bidirectional long short-term memory (Bi-LSTM) network to learn representations from temporally sampled signals. The second stream extracts latent representations from the PPG time-frequency spectrogram using a five-layer CNN. The outputs from both streams are fed into a fusion layer for the outcome. Visualization of the attention weights learned demonstrates the effectiveness of the attention mechanism against noise. The experimental results show that the TSHNN outperforms all the competitive baseline approaches and with 98.09% accuracy, achieves state-of-the-art performance.

Keywords: PPG wearables, atrial fibrillation, feature fusion, attention mechanism, hyber network

Procedia PDF Downloads 93

Authors: Chongya Song, Kang Yen, Alexander Pons, Jin Liu

Abstract:

In the field of machine learning, the ensemble has been employed as a common methodology to improve the performance upon multiple base classifiers. However, the true predictions are often canceled out by the false ones during consensus due to a phenomenon called “curse of correlation” which is represented as the strong interferences among the predictions produced by the base classifiers. In addition, the existing practices are still not able to effectively mitigate the problem of imbalanced classification. Based on the analysis on our experiment results, we conclude that the two problems are caused by some inherent deficiencies in the approach of consensus. Therefore, we create an enhanced ensemble algorithm which adopts a designed rank-based chain-mode consensus to overcome the two problems. In order to evaluate the proposed ensemble algorithm, we employ a well-known benchmark data set NSL-KDD (the improved version of dataset KDDCup99 produced by University of New Brunswick) to make comparisons between the proposed and 8 common ensemble algorithms. Particularly, each compared ensemble classifier uses the same 22 base classifiers, so that the differences in terms of the improvements toward the accuracy and reliability upon the base classifiers can be truly revealed. As a result, the proposed rank-based chain-mode consensus is proved to be a more effective ensemble solution than the traditional consensus approach, which outperforms the 8 ensemble algorithms by 20% on almost all compared metrices which include accuracy, precision, recall, F1-score and area under receiver operating characteristic curve.

Keywords: consensus, curse of correlation, imbalance classification, rank-based chain-mode ensemble

Procedia PDF Downloads 113

Authors: Larry Powell, Seth Polsley, Drew Casey, Tracy Hammond

Abstract:

Human activity recognition (HAR) systems have shown positive performance when recognizing repetitive activities like walking, running, and sleeping. Water-based activities are a reasonably new area for activity recognition. However, water-based activity recognition has largely focused on supporting the elite and competitive swimming population, which already has amazing coordination and proper form. Beginner swimmers are not perfect, and activity recognition needs to support the individual motions to help beginners. Activity recognition algorithms are traditionally built around short segments of timed sensor data. Using a time window input can cause performance issues in the machine learning model. The window’s size can be too small or large, requiring careful tuning and precise data segmentation. In this work, we present a method that uses a time window as the initial segmentation, then separates the data based on the change in the sensor value. Our system uses a multi-phase segmentation method that pulls all peaks and valleys for each axis of an accelerometer placed on the swimmer’s lower back. This results in high recognition performance using leave-one-subject-out validation on our study with 20 beginner swimmers, with our model optimized from our final dataset resulting in an F-Score of 0.95.

Keywords: time window, peak/valley segmentation, feature extraction, beginner swimming, activity recognition

Procedia PDF Downloads 103

Authors: Wenjun Hou, Marek Perkowski

Abstract:

The Traveling Salesman problem is famous in computing and graph theory. In short, it asks for the Hamiltonian cycle of the least total weight in a given graph with N nodes. All variations on this problem, such as those with K-bounded-degree nodes, are classified as NP-complete in classical computing. Although several papers propose theoretical high-level designs of quantum algorithms for the Traveling Salesman Problem, no quantum circuit implementation of these algorithms has been created up to our best knowledge. In contrast to previous papers, the goal of this paper is not to optimize some abstract complexity measures based on the number of oracle iterations, but to be able to evaluate the real circuit and time costs of the quantum computer. Using the emerging quantum programming language Q# developed by Microsoft, which runs quantum circuits in a quantum computer simulation, an implementation of the bounded-degree problem and its respective quantum circuit were created. To apply Grover’s algorithm to this problem, a quantum oracle was designed, evaluating the cost of a particular set of edges in the graph as well as its validity as a Hamiltonian cycle. Repeating the Grover algorithm with an oracle that finds successively lower cost each time allows to transform the decision problem to an optimization problem, finding the minimum cost of Hamiltonian cycles. N log₂ K qubits are put into an equiprobablistic superposition by applying the Hadamard gate on each qubit. Within these N log₂ K qubits, the method uses an encoding in which every node is mapped to a set of its encoded edges. The oracle consists of several blocks of circuits: a custom-written edge weight adder, node index calculator, uniqueness checker, and comparator, which were all created using only quantum Toffoli gates, including its special forms, which are Feynman and Pauli X. The oracle begins by using the edge encodings specified by the qubits to calculate each node that this path visits and adding up the edge weights along the way. Next, the oracle uses the calculated nodes from the previous step and check that all the nodes are unique. Finally, the oracle checks that the calculated cost is less than the previously-calculated cost. By performing the oracle an optimal number of times, a correct answer can be generated with very high probability. The oracle of the Grover Algorithm is modified using the recalculated minimum cost value, and this procedure is repeated until the cost cannot be further reduced. This algorithm and circuit design have been verified, using several datasets, to generate correct outputs.

Keywords: quantum computing, quantum circuit optimization, quantum algorithms, hybrid quantum algorithms, quantum programming, Grover’s algorithm, traveling salesman problem, bounded-degree TSP, minimal cost, Q# language

Procedia PDF Downloads 167

Authors: Simone Leonardi, Luca Ardito

Abstract:

Micro-influencers are effective elements in the marketing strategies of companies and institutions because of their capability to create an hyper-engaged audience around a specific topic of interest. In recent years, many scientific approaches and commercial tools have handled the task of detecting this type of social media users. These strategies adopt solutions ranging from rule based machine learning models to deep neural networks and graph analysis on text, images, and account information. This work compares the existing solutions and proposes an ensemble method to generalize them with different input data and social media platforms. The deployed solution combines deep learning models on unstructured data with statistical machine learning models on structured data. We retrieve both social media accounts information and multimedia posts on Twitter and Instagram. These data are mapped into feature vectors for an eXtreme Gradient Boosting (XGBoost) classifier. Sixty different topics have been analyzed to build a rule based gold standard dataset and to compare the performances of our approach against baseline classifiers. We prove the effectiveness of our work by comparing the accuracy, precision, recall, and f1 score of our model with different configurations and architectures. We obtained an accuracy of 0.91 with our best performing model.

Keywords: deep learning, gradient boosting, image processing, micro-influencers, NLP, social media

Procedia PDF Downloads 154

Authors: Dan Geneau, Mary Claire Geneau, Seth Lenetsky, Ming -Chang Tsai, Marc Klimstra

Abstract:

Introduction— Horizontal force-velocity profiling (HFVP) involves modeling an athletes linear sprint kinematics to estimate valuable maximum force and velocity metrics. This approach to performance modeling has been used in field-based team sports and has recently been introduced to ice-hockey as a forward skating performance assessment. While preliminary data has been collected on ice, distance constraints of the on-ice test restrict the ability of the athletes to reach their maximal velocity which result in limits of the model to effectively estimate athlete performance. This is especially true of more elite athletes. This report explores whether athletes on-ice are able to reach a velocity plateau similar to what has been seen in overground trials. Fourteen male Major Junior ice-hockey players (BW= 83.87 +/- 7.30 kg, height = 188 ± 3.4cm cm, age = 18 ± 1.2 years n = 14) were recruited. For on-ice sprints, participants completed a standardized warm-up consisting of skating and dynamic stretching and a progression of three skating efforts from 50% to 95%. Following the warm-up, participants completed three on ice 45m sprints, with three minutes of rest in between each trial. For overground sprints, participants completed a similar dynamic warm-up to that of on-ice trials. Following the warm-up participants completed three 40m overground sprint trials. For each trial (on-ice and overground), radar was used to collect instantaneous velocity (Stalker ATS II, Texas, USA) aimed at the participant’s waist. Sprint velocities were modelled using custom Python (version 3.2) script using a mono-exponential function, similar to previous work. To determine if on-ice tirals were achieving a maximum velocity (plateau), minimum acceleration values of the modeled data at the end of the sprint were compared (using paired t-test) between on-ice and overground trials. Significant differences (P<0.001) between overground and on-ice minimum accelerations were observed. It was found that on-ice trials consistently reported higher final acceleration values, indicating a maximum maintained velocity (plateau) had not been reached. Based on these preliminary findings, it is suggested that reliable HFVP metrics cannot yet be collected from all ice-hockey populations using current methods. Elite male populations were not able to achieve a velocity plateau similar to what has been seen in overground trials, indicating the absence of a maximum velocity measure. With current velocity and acceleration modeling techniques, including a dependency of a velocity plateau, these results indicate the potential for error in on-ice HFVP measures. Therefore, these findings suggest that a greater on-ice sprint distance may be required or the need for other velocity modeling techniques, where maximal velocity is not required for a complete profile.   

Keywords: ice-hockey, sprint, skating, power

Procedia PDF Downloads 84

Authors: Saqib Rasool, Adnan N. Mian

Abstract:

Web components are introduced as the latest standard of HTML5 for writing modular web interfaces for ensuring maintainability through the isolated scope of web components. Reusability can also be achieved by sharing plug-and-play web components that can be used as off-the-shelf components by other developers. A web component encapsulates all the required HTML, CSS and JavaScript code as a standalone package which must be imported for integrating a web component within an existing web interface. It is then followed by the integration of web component with the web services for dynamically populating its content. Since web components are reusable as off-the-shelf components, these must be equipped with some mechanism for ensuring their proper integration with web services. The consistency of a service behavior can be verified through type-checking. This is one of the popular solutions for improving the quality of code in many programming languages. However, HTML does not provide type checking as it is a markup language and not a programming language. The contribution of this work is to introduce a new extension of HTML called Service Type-checking Markup Language (STML) for adding support of type checking in HTML for JSON based REST services. STML can be used for defining the expected data types of response from JSON based REST services which will be used for populating the content within HTML elements of a web component. Although JSON has five data types viz. string, number, boolean, object and array but STML is made to supports only string, number and object. This is because of the fact that both object and array are considered as string, when populated in HTML elements. In order to define the data type of any HTML element, developer just needs to add the custom STML attributes of st-string, st-number and st-boolean for string, number and boolean respectively. These all annotations of STML are used by the developer who is writing a web component and it enables the other developers to use automated type-checking for ensuring the proper integration of their REST services with the same web component. Two utilities have been written for developers who are using STML based web components. One of these utilities is used for automated type-checking during the development phase. It uses the browser console for showing the error description if integrated web service is not returning the response with expected data type. The other utility is a Gulp based command line utility for removing the STML attributes before going in production. This ensures the delivery of STML free web pages in the production environment. Both of these utilities have been tested to perform type checking of REST services through STML based web components and results have confirmed the feasibility of evaluating service behavior only through HTML. Currently, STML is designed for automated type-checking of integrated REST services but it can be extended to introduce a complete service testing suite based on HTML only, and it will transform STML from Service Type-checking Markup Language to Service Testing Markup Language.

Keywords: REST, STML, type checking, web component

Procedia PDF Downloads 233

Authors: Kishore Kumar R., Kaustav Sengupta, Shalini Sood Sehgal, Poornima Santhanam

Abstract:

Fashion is a human expression that is constantly changing. One of the prime factors that consistently influences fashion is the change in colour preferences. The role of colour in our everyday lives is very significant. It subconsciously explains a lot about one’s mindset and mood. Analyzing the colours by extracting them from the outfit images is a critical study to examine the individual’s/consumer behaviour. Several research works have been carried out on extracting colours from images, but to the best of our knowledge, there were no studies that extract colours to specific apparel and identify colour patterns geographically. This paper proposes a framework for accurately extracting colours from T-shirt images and predicting dominant colours geographically. The proposed method consists of two stages: first, a U-Net deep learning model is adopted to segment the T-shirts from the images. Second, the colours are extracted only from the T-shirt segments. The proposed method employs the iMaterialist (Fashion) 2019 dataset for the semantic segmentation task. The proposed framework also includes a mechanism for gathering data and analyzing India’s general colour preferences. From this research, it was observed that black and grey are the dominant colour in different regions of India. The proposed method can be adapted to study fashion’s evolving colour preferences.

Keywords: colour analysis in t-shirts, convolutional neural network, encoder-decoder, k-means clustering, semantic segmentation, U-Net model

Procedia PDF Downloads 86

Authors: Mais Nijim, Rama Devi Chennuboyina, Waseem Al Aqqad

Abstract:

Advances in spatial and spectral resolution of satellite images have led to tremendous growth in large image databases. The data we acquire through satellites, radars and sensors consists of important geographical information that can be used for remote sensing applications such as region planning, disaster management. Spatial data classification and object recognition are important tasks for many applications. However, classifying objects and identifying them manually from images is a difficult task. Object recognition is often considered as a classification problem, this task can be performed using machine-learning techniques. Despite of many machine-learning algorithms, the classification is done using supervised classifiers such as Support Vector Machines (SVM) as the area of interest is known. We proposed a classification method, which considers neighboring pixels in a region for feature extraction and it evaluates classifications precisely according to neighboring classes for semantic interpretation of region of interest (ROI). A dataset has been created for training and testing purpose; we generated the attributes by considering pixel intensity values and mean values of reflectance. We demonstrated the benefits of using knowledge discovery and data-mining techniques, which can be on image data for accurate information extraction and classification from high spatial resolution remote sensing imagery.

Keywords: remote sensing, object recognition, classification, data mining, waterbody identification, feature extraction

Procedia PDF Downloads 319

Authors: L. Yu, W. K. Li, S. K. Ong, A. Y. C. Nee

Abstract:

In this paper, a scalable augmented reality framework for handheld devices is presented. The presented framework is enabled by using a server-client data communication structure, in which the search for tracking targets among a database of images is performed on the server-side while pixel-wise 3D tracking is performed on the client-side, which, in this case, is a handheld mobile device. Image search on the server-side adopts a residual-enhanced image descriptors representation that gives the framework a scalability property. The tracking algorithm on the client-side is based on a gravity-aligned feature descriptor which takes the advantage of a sensor-equipped mobile device and an optimized intensity-based image alignment approach that ensures the accuracy of 3D tracking. Automatic content streaming is achieved by using a key-frame selection algorithm, client working phase monitoring and standardized rules for content communication between the server and client. The recognition accuracy test performed on a standard dataset shows that the method adopted in the presented framework outperforms the Bag-of-Words (BoW) method that has been used in some of the previous systems. Experimental test conducted on a set of video sequences indicated the real-time performance of the tracking system with a frame rate at 15-30 frames per second. The presented framework is exposed to be functional in practical situations with a demonstration application on a campus walk-around.

Keywords: augmented reality framework, server-client model, vision-based tracking, image search

Procedia PDF Downloads 262

Authors: Jack R. McKenzie, Peter A. Appleby, Thomas House, Neil Walton

Abstract:

Cold-start is a notoriously difficult problem which can occur in recommendation systems, and arises when there is insufficient information to draw inferences for users or items. To address this challenge, a contextual bandit algorithm – the Fast Approximate Bayesian Contextual Cold Start Learning algorithm (FAB-COST) – is proposed, which is designed to provide improved accuracy compared to the traditionally used Laplace approximation in the logistic contextual bandit, while controlling both algorithmic complexity and computational cost. To this end, FAB-COST uses a combination of two moment projection variational methods: Expectation Propagation (EP), which performs well at the cold start, but becomes slow as the amount of data increases; and Assumed Density Filtering (ADF), which has slower growth of computational cost with data size but requires more data to obtain an acceptable level of accuracy. By switching from EP to ADF when the dataset becomes large, it is able to exploit their complementary strengths. The empirical justification for FAB-COST is presented, and systematically compared to other approaches on simulated data. In a benchmark against the Laplace approximation on real data consisting of over 670, 000 impressions from autotrader.co.uk, FAB-COST demonstrates at one point increase of over 16% in user clicks. On the basis of these results, it is argued that FAB-COST is likely to be an attractive approach to cold-start recommendation systems in a variety of contexts.

Keywords: cold-start learning, expectation propagation, multi-armed bandits, Thompson Sampling, variational inference

Procedia PDF Downloads 95

Authors: Pattra Pukdeekiat, Siriluk Ruangrungrote

Abstract:

This work was originated to precisely determine direct solar radiation by using atmospheric physics models since the accurate prediction of solar radiation is necessary and useful for solar energy applications including atmospheric research. The possible models and techniques for a calculation of regional direct solar radiation were challenging and compulsory for the case of unavailable instrumental measurement. The investigation was mathematically governed by six astronomical parameters i.e. declination (δ), hour angle (ω), solar time, solar zenith angle (θz), extraterrestrial radiation (Iso) and eccentricity (E0) along with two atmospheric parameters i.e. air mass (mr) and dew point temperature at Bangna meteorological station (13.67° N, 100.61° E) in Bangkok, Thailand. Analyses of five models of solar radiation determination with the assumption of clear sky were applied accompanied by three statistical tests: Mean Bias Difference (MBD), Root Mean Square Difference (RMSD) and Coefficient of determination (R2) in order to validate the accuracy of obtainable results. The calculated direct solar radiation was in a range of 491-505 Watt/m2 with relative percentage error 8.41% for winter and 532-540 Watt/m2 with relative percentage error 4.89% for summer 2014. Additionally, dataset of seven continuous days, representing both seasons were considered with the MBD, RMSD and R2 of -0.08, 0.25, 0.86 and -0.14, 0.35, 3.29, respectively, which belong to Kumar model for winter and CSR model for summer. In summary, the determination of direct solar radiation based on atmospheric models and empirical equations could advantageously provide immediate and reliable values of the solar components for any site in the region without a constraint of actual measurement.

Keywords: atmospheric physics models, astronomical parameters, atmospheric parameters, clear sky condition

Procedia PDF Downloads 391

Authors: A. M. John-Otumu, M. M. Rahman, M. C. Onuoha, E. P. Ojonugwa

Abstract:

A Convolutional Neural Network (CNN) model for predicting Lassa fever was built using Python 3.8.0 programming language, alongside Keras 2.2.4 and TensorFlow 2.6.1 libraries as the development environment in order to reduce the current high risk of Lassa fever in West Africa, particularly in Nigeria. The study was prompted by some major flaws in existing conventional laboratory equipment for diagnosing Lassa fever (RT-PCR), as well as flaws in AI-based techniques that have been used for probing and prognosis of Lassa fever based on literature. There were 15,679 blood smear microscopic image datasets collected in total. The proposed model was trained on 70% of the dataset and tested on 30% of the microscopic images in avoid overfitting. A 3x3x3 convolution filter was also used in the proposed system to extract features from microscopic images. The proposed CNN-based model had a recall value of 96%, a precision value of 93%, an F1 score of 95%, and an accuracy of 94% in predicting and accurately classifying the images into clean or infected samples. Based on empirical evidence from the results of the literature consulted, the proposed model outperformed other existing AI-based techniques evaluated. If properly deployed, the model will assist physicians, medical laboratory scientists, and patients in making accurate diagnoses for Lassa fever cases, allowing the mortality rate due to the Lassa fever virus to be reduced through sound decision-making.

Keywords: artificial intelligence, ANN, blood smear, CNN, deep learning, Lassa fever

Procedia PDF Downloads 89

Authors: Benkouiten Imene, Chabli Ouerdia, Boutoutaou Hamid, Kadri Nesrine, Bouledroua Omar

Abstract:

Algeria is actively enhancing gas productivity by augmenting the supply flow. However, this effort has led to increased internal pressure, posing a potential risk to the pipeline's integrity, particularly in the presence of corrosion defects. Sonatrach relies on a vast network of pipelines spanning 24,000 kilometers for the transportation of gas and oil. The aging of these pipelines raises the likelihood of corrosion both internally and externally, heightening the risk of ruptures. To address this issue, a comprehensive inspection is imperative, utilizing specialized scraping tools. These advanced tools furnish a detailed assessment of all pipeline defects. It is essential to recalculate the pressure parameters to safeguard the corroded pipeline's integrity while ensuring the continuity of production. In this context, Sonatrach employs symbolic pressure limit calculations, such as ASME B31G (2009) and the modified ASME B31G (2012). The aim of this study is to perform a comparative analysis of various limit pressure calculation methods documented in the literature, namely DNV RP F-101, SHELL, P-CORRC, NETTO, and CSA Z662. This comparative assessment will be based on a dataset comprising 329 burst tests published in the literature. Ultimately, we intend to introduce a novel approach grounded in the finite element method, employing ANSYS software.

Keywords: pipeline burst pressure, burst test, corrosion defect, corroded pipeline, finite element method

Procedia PDF Downloads 43

Authors: Tae-Uk Yun, Pullip Chung, Kee-Young Kwahk, Hyunchul Ahn

Abstract:

Fake news is defined as the news articles that are intentionally and verifiably false, and could mislead readers. Spread of fake news may provoke anxiety, chaos, fear, or irrational decisions of the public. Thus, detecting fake news and preventing its spread has become very important issue in our society. However, due to the huge amount of fake news produced every day, it is almost impossible to identify it by a human. Under this context, researchers have tried to develop automated fake news detection using machine learning techniques over the past years. But, there have been no prior studies proposed an automated fake news detection method for Korean news to our best knowledge. In this study, we aim to detect Korean fake news using text mining and machine learning techniques. Our proposed method consists of two steps. In the first step, the news contents to be analyzed is convert to quantified values using various text mining techniques (topic modeling, TF-IDF, and so on). After that, in step 2, classifiers are trained using the values produced in step 1. As the classifiers, machine learning techniques such as logistic regression, backpropagation network, support vector machine, and deep neural network can be applied. To validate the effectiveness of the proposed method, we collected about 200 short Korean news from Seoul National University’s FactCheck. which provides with detailed analysis reports from 20 media outlets and links to source documents for each case. Using this dataset, we will identify which text features are important as well as which classifiers are effective in detecting Korean fake news.

Keywords: fake news detection, Korean news, machine learning, text mining

Procedia PDF Downloads 248

Authors: C. L. Huang, J. J. Sia

Abstract:

The interest-based social networks, also known as social bookmark sharing systems, are useful platforms for people to conveniently read and collect internet resources. These platforms also providing function of social networks, and users can share and explore internet resources from the social networks. Providing personalized internet resources to users is an important issue on these platforms. This study uses two types of relationship on the social networks—following and follower and proposes a collaborative recommender system, consisting of two main steps. First, this study calculates the relationship strength between the target user and the target user's followings and followers to find top-N similar neighbors. Second, from the top-N similar neighbors, the articles (internet resources) that may interest the target user are recommended to the target user. In this system, users can efficiently obtain recent, related and diverse internet resources (knowledge) from the interest-based social network. This study collected the experimental dataset from Diigo, which is a famous bookmark sharing system. The experimental results show that the proposed recommendation model is more accurate than two traditional baseline recommendation models but slightly lower than the cosine model in accuracy. However, in the metrics of the diversity and executing time, our proposed model outperforms the cosine model.

Keywords: recommender systems, social networks, tagging, bookmark sharing systems, collaborative recommender systems, knowledge management

Procedia PDF Downloads 149

Authors: Kolade Olawande Owoeye, George Weir

Abstract:

Terrorism and extremism activities on the internet are becoming the most significant threats to national security because of their potential dangers. In response to this challenge, law enforcement and security authorities are actively implementing comprehensive measures by countering the use of the internet for terrorism. To achieve the measures, there is need for intelligence gathering via the internet. This includes real-time monitoring of potential websites that are used for recruitment and information dissemination among other operations by extremist groups. However, with billions of active webpages, real-time monitoring of all webpages become almost impossible. To narrow down the search domain, there is a need for efficient webpage classification techniques. This research proposed a new approach tagged: SentiPosit-based method. SentiPosit-based method combines features of the Posit-based method and the Sentistrenght-based method for classification of terrorism and extremism webpages. The experiment was carried out on 7500 webpages obtained through TENE-webcrawler by International Cyber Crime Research Centre (ICCRC). The webpages were manually grouped into three classes which include the ‘pro-extremist’, ‘anti-extremist’ and ‘neutral’ with 2500 webpages in each category. A supervised learning algorithm is then applied on the classified dataset in order to build the model. Results obtained was compared with existing classification method using the prediction accuracy and runtime. It was observed that our proposed hybrid approach produced a better classification accuracy compared to existing approaches within a reasonable runtime.

Keywords: sentiposit, classification, extremism, terrorism

Procedia PDF Downloads 256

Authors: S. Hutasavi, D. Chen

Abstract:

The built-up area is a significant proxy to measure regional economic growth and reflects the Gross Provincial Product (GPP). However, an up-to-date and reliable database of built-up areas is not always available, especially in developing countries. The cloud-based geospatial analysis platform such as Google Earth Engine (GEE) provides an opportunity with accessibility and computational power for those countries to generate the built-up data. Therefore, this study aims to extract the built-up areas in Eastern Economic Corridor (EEC), Thailand using day and nighttime satellite imagery based on GEE facilities. The normalized indices were generated from Landsat 8 surface reflectance dataset, including Normalized Difference Built-up Index (NDBI), Built-up Index (BUI), and Modified Built-up Index (MBUI). These indices were applied to identify built-up areas in EEC. The result shows that MBUI performs better than BUI and NDBI, with the highest accuracy of 0.85 and Kappa of 0.82. Moreover, the overall accuracy of classification was improved from 79% to 90%, and error of total built-up area was decreased from 29% to 0.7%, after night-time light data from the Visible and Infrared Imaging Suite (VIIRS) Day Night Band (DNB). The results suggest that MBUI with night-time light imagery is appropriate for built-up area extraction and be utilize for further study of socioeconomic impacts of regional development policy over the EEC region.

Keywords: built-up area extraction, google earth engine, adaptive thresholding method, rapid mapping

Procedia PDF Downloads 106