Search results for: train

Authors: Ajoy Kumar Das, Prasenjit Dey

Abstract:

Heat transfer due to forced convection of copper water based nanofluid has been predicted by Artificial Neural network (ANN). The present nanofluid is formed by mixing copper nano particles in water and the volume fractions are considered here are 0% to 15% and the Reynolds number are kept constant at 100. The back propagation algorithm is used to train the network. The present ANN is trained by the input and output data which has been obtained from the numerical simulation, performed in finite volume based Computational Fluid Dynamics (CFD) commercial software Ansys Fluent. The numerical simulation based results are compared with the back propagation based ANN results. It is found that the forced convection heat transfer of water based nanofluid can be predicted correctly by ANN. It is also observed that the back propagation ANN can predict the heat transfer characteristics of nanofluid very quickly compared to standard CFD method.

Keywords: forced convection, square cylinder, nanofluid, neural network

Procedia PDF Downloads 294

Authors: Ci Lin, Tet Yeap, Iluju Kiringa

Abstract:

This paper introduces the subspace rotation algorithm (SRA) to train the Restricted Hopfield Network (RHN) as an auto-associative memory. Subspace rotation algorithm is a gradient-free subspace tracking approach based on the singular value decomposition (SVD). In comparison with Backpropagation Through Time (BPTT) on training RHN, it is observed that SRA could always converge to the optimal solution and BPTT could not achieve the same performance when the model becomes complex, and the number of patterns is large. The AUTS case study showed that the RHN model trained by SRA could achieve a better structure of attraction basin with larger radius(in general) than the Hopfield Network(HNN) model trained by Hebbian learning rule. Through learning 10000 patterns from MNIST dataset with RHN models with different number of hidden nodes, it is observed that an several components could be adjusted to achieve a balance between recovery accuracy and noise resistance.

Keywords: hopfield neural network, restricted hopfield network, subspace rotation algorithm, hebbian learning rule

Procedia PDF Downloads 85

Authors: Yang Zhang, Jian He

Abstract:

Heart disease is one of the leading causes of death in the world, and coronary heart disease (CHD) is one of the major heart diseases. Electrocardiogram (ECG) is widely used in the detection of heart diseases, but the traditional manual method for CHD prediction by analyzing ECG requires lots of professional knowledge for doctors. This paper introduces sliding window and continuous wavelet transform (CWT) to transform ECG signals into images, and then ResNet and Bi-LSTM are introduced to build the ECG feature extraction network (namely ECGNet). At last, an auxiliary system for coronary heart disease prediction was developed based on modified ResNet18 and Bi-LSTM, and the public ECG dataset of CHD from MIMIC-3 was used to train and test the system. The experimental results show that the accuracy of the method is 83%, and the F1-score is 83%. Compared with the available methods for CHD prediction based on ECG, such as kNN, decision tree, VGGNet, etc., this method not only improves the prediction accuracy but also could avoid the degradation phenomenon of the deep learning network.

Keywords: Bi-LSTM, CHD, ECG, ResNet, sliding window

Procedia PDF Downloads 50

Authors: Bulus Kpame, Mukhtar Mohammed Alhaji, Garba Jibril

Abstract:

In any team sport, the most important variables that should be used to measure performance are physical condition, and technical and tactical performance. In a complex game like football, it is extremely difficult to measure the relative importance of each of these variables. However, physical fitness itself has been shown to consist of several components, like endurance, strength, flexibility, agility, coordination and speed. Each of these components has been shown to consist of several subcomponents. This paper attempts to describe a test battery to assess and evaluate physical performance in football players. This battery comprises a functional, structured training session of about 2.5hrs. it consists of quality rating of the warm-up procedure, tests of flexibility, football skills, power, speed, and endurance. Acceptable values for performance in each of the tests are also presented under each test. It is hoped that this battery of tests will be helpful to the coach in determining the effect of a specific training program. It would also be helpful to train physician and trainer, to monitor progress during rehabilitation after sustaining any injury.

Keywords: assessment, evaluation, performance, programs

Procedia PDF Downloads 368

Authors: Cyril Kaplan, Nikola Jajcay

Abstract:

We developed purely sonic, musical based, highly customizable EEG neurofeedback environment designed to administer a new neurofeedback training protocol. The training protocol concentrates on improving the ability to switch between several mental states characterized by different levels of arousal, each of them correlated to specific brain wave activity patterns in several specific regions of neocortex. This paper describes the neurofeedback training environment we developed and its specificities, thus can be helpful as a manual to guide other neurofeedback users (both researchers and practitioners) interested in our editable open source program (available to download and usage under CC license). Responses and reaction of first trainees that used our environment are presented in this article. Combination of qualitative methods (thematic analysis of neurophenomenological insights of trainees and post-session semi-structured interviews) and quantitative methods (power spectra analysis of EEG recorded during the training) were employed to obtain a multifaceted view on our new training protocol.

Keywords: EEG neurofeedback, mixed methods, self-regulation, switch-between-states training

Procedia PDF Downloads 183

Authors: Christoph Linse, Thomas Martinetz

Abstract:

Recent findings have shown that Neural Networks generalize also in over-parametrized regimes with zero training error. This is surprising, since it is completely against traditional machine learning wisdom. In our empirical study we fortify these findings in the domain of fine-grained image classification. We show that very large Convolutional Neural Networks with millions of weights do learn with only a handful of training samples and without image augmentation, explicit regularization or pretraining. We train the architectures ResNet018, ResNet101 and VGG19 on subsets of the difficult benchmark datasets Caltech101, CUB_200_2011, FGVCAircraft, Flowers102 and StanfordCars with 100 classes and more, perform a comprehensive comparative study and draw implications for the practical application of CNNs. Finally, we show that VGG19 with 140 million weights learns to distinguish airplanes and motorbikes with up to 95% accuracy using only 20 training samples per class.

Keywords: convolutional neural networks, fine-grained image classification, generalization, image recognition, over-parameterized, small data sets

Procedia PDF Downloads 54

Authors: Tanish Mittal, Preyansh Agrawal, Esha Pahwa, Aarya Makwana

Abstract:

Multi-Scale and U-shaped Networks are widely used in various image restoration problems, including deblurring. Keeping in mind the wide range of applications, we present a comparison of these architectures and their effects on image deblurring. We also introduce a new block called as NFResblock. It consists of a Fast Fourier Transformation layer and a series of modified Non-Linear Activation Free Blocks. Based on these architectures and additions, we introduce NFResnet and NFResnet+, which are modified multi-scale and U-Net architectures, respectively. We also use three differ-ent loss functions to train these architectures: Charbonnier Loss, Edge Loss, and Frequency Reconstruction Loss. Extensive experiments on the Deep Video Deblurring dataset, along with ablation studies for each component, have been presented in this paper. The proposed architectures achieve a considerable increase in Peak Signal to Noise (PSNR) ratio and Structural Similarity Index (SSIM) value.

Keywords: multi-scale, Unet, deblurring, FFT, resblock, NAF-block, nfresnet, charbonnier, edge, frequency reconstruction

Procedia PDF Downloads 89

Authors: Shehret Tilvaldyev, Jorge Flores-Garay, Alfredo Villanueva, Erwin Martinez, Lazaro Rico

Abstract:

The efficiency of modern transportation is severely compromised by the prevalence of turbulent drag. The high level of turbulent skin-friction occurring, e.g., on the surface of an aircraft, automobiles or the carriage of a high-speed train, is responsible for excess fuel consumption and increased carbon emissions. The environmental, political, and economic pressure to improve fuel efficiency and reduce carbon emissions associated with transportation means that reducing turbulent skin-friction drag is a pressing engineering problem. The dynamic pressure of subsonic airflow around solid objects creates lift, but also induces drag force. This paper is presenting the results of laboratory experiments, investigating appearance and magnitude of dynamic pressure in micro scale of subsonic air flow around right cylinder and symmetrical airfoil.

Keywords: airflow, dynamic pressure, micro scale, symmetric object

Procedia PDF Downloads 351

Authors: Ting (Grace) Lin, Jianhong (Cecilia) Xia, Todd Robinson

Abstract:

Accessibility analysis, examining people’s ability to access facilities and destinations, is a fundamental assessment for transport planning, policy making, and social exclusion research. Dynamic accessibility which measures accessibility in real-time traffic environment has been an advanced accessibility indicator in transport research. It is also a useful indicator to help travelers to understand travel time daily variability, assists traffic engineers to monitor traffic congestions, and finally develop effective strategies in order to mitigate traffic congestions. This research involved real-time traffic information by collecting travel time data with 15-minute interval via the TomTom^® API. A framework for measuring dynamic accessibility was then developed based on the gravity theory and accessibility dichotomy theory through space and time interpolation. Finally, the dynamic accessibility can be derived at any given time and location under dynamic accessibility spatial analysis framework.

Keywords: dynamic accessibility, hot spot, transport research, TomTom® API

Procedia PDF Downloads 353

Authors: Sadeem A. Alolayan

Abstract:

The purpose of early intervention (EI) programs and services is to minimize the impact of disability on children ages 0-5 and to reduce future special education costs. This literature review identifies the status of families of children with special needs. Four major themes emerged from this literature review. The first was the family’s needs and the expressed desire for services to be obtained or outcomes to be achieved. The second was family support, meaning any information or skills needed to facilitate parents’ role as professionals in order to enable them to train and provide their child with the best quality of life. The third theme, barriers, was defined as parents’ actions or life circumstances that hindered families in obtaining appropriate EI services. The conclusions derived from the recommendations are that effective parent participation involves careful planning, establishing and maintaining a trusted rapport between parents, and EI providers that understand parents’ individual needs and interests, thus motivating effective parent involvement in early intervention programs.

Keywords: early intervention, individuals with disabilities education act, parents, recommendations

Procedia PDF Downloads 144

Authors: Hyeongbok Kim, Lingling Zhao, Xiaohong Su

Abstract:

Deep reinforcement learning has been applied to address various problems in robotics, such as autonomous driving and unmanned aerial vehicle. However, because of the sparse reward penalty for a collision with obstacles during the navigation mission, the agent fails to learn the optimal policy or requires a long time for convergence. Therefore, using obstacles and enemy agents, in this paper, we present a curriculum-based boost learning method to effectively train compound skills during multi-agent reinforcement learning. First, to enable the agents to solve challenging tasks, we gradually increased learning difficulties by adjusting reward shaping instead of constructing different learning environments. Then, in a benchmark environment with static obstacles and moving enemy agents, the experimental results showed that the proposed curriculum learning strategy enhanced cooperative navigation and compound collision avoidance skills in uncertain environments while improving learning efficiency.

Keywords: curriculum learning, hard exploration, multi-agent reinforcement learning, robotic navigation, sparse reward

Procedia PDF Downloads 62

Authors: Wichian Sriprachan, Chutikarn Sriviboon

Abstract:

The objectives of this research were to study the management of local towns and to develop a better model of town management according to the Philosophy of Sufficiency Economy. This study utilized qualitative research, field research, as well as documentary research at the same time. A total of 10 local towns or Tambons of Supanburi province, Thailand were selected for an in-depth interview. The findings revealed that the model of local town management according to Philosophy of Sufficient Economy was in a level of “good” and the model of management has the five basic guidelines: 1) ability to manage budget information and keep it up-to-date, 2) ability to decision making according to democracy rules, 3) ability to use check and balance system, 4) ability to control, follow, and evaluation, and 5) ability to allow the general public to participate. In addition, the findings also revealed that the human resource management according to Philosophy of Sufficient Economy includes obeying laws, using proper knowledge, and having integrity in five areas: plan, recruit, select, train, and maintain human resources.

Keywords: management, local town (Tambon), principles of sufficiency economy, marketing management

Procedia PDF Downloads 314

Authors: Yihao Kuang, Bowen Ding

Abstract:

With the increasing scale and complexity of knowledge graph, modern knowledge graph contains more and more types of entity, relationship, and attribute information. Therefore, in recent years, it has been a trend for knowledge graph inference to use reinforcement learning to deal with large-scale, incomplete, and noisy knowledge graphs and improve the inference effect and interpretability. The Proximal Policy Optimization (PPO) algorithm utilizes a near-end strategy optimization approach. This allows for more extensive updates of policy parameters while constraining the update extent to maintain training stability. This characteristic enables PPOs to converge to improved strategies more rapidly, often demonstrating enhanced performance early in the training process. Furthermore, PPO has the advantage of offline learning, effectively utilizing historical experience data for training and enhancing sample utilization. This means that even with limited resources, PPOs can efficiently train for reinforcement learning tasks. Based on these characteristics, this paper aims to obtain a better and more efficient inference effect by introducing PPO into knowledge inference technology.

Keywords: reinforcement learning, PPO, knowledge inference

Procedia PDF Downloads 193

Authors: Bitewulign Mekonnen

Abstract:

Context: This scientific paper focuses on the use of near-infrared (NIR) spectroscopy to determine glucose concentration in aqueous solutions accurately and rapidly. The study compares six different machine learning methods for predicting glucose concentration and also explores the development of a deep learning model for classifying NIR spectra. The objective is to optimize the detection model and improve the accuracy of glucose prediction. This research is important because it provides a comprehensive analysis of various machine-learning techniques for estimating aqueous glucose concentrations. Research Aim: The aim of this study is to compare and evaluate different machine-learning methods for predicting glucose concentration from NIR spectra. Additionally, the study aims to develop and assess a deep-learning model for classifying NIR spectra. Methodology: The research methodology involves the use of machine learning and deep learning techniques. Six machine learning regression models, including support vector machine regression, partial least squares regression, extra tree regression, random forest regression, extreme gradient boosting, and principal component analysis-neural network, are employed to predict glucose concentration. The NIR spectra data is randomly divided into train and test sets, and the process is repeated ten times to increase generalization ability. In addition, a convolutional neural network is developed for classifying NIR spectra. Findings: The study reveals that the SVMR, ETR, and PCA-NN models exhibit excellent performance in predicting glucose concentration, with correlation coefficients (R) > 0.99 and determination coefficients (R²)> 0.985. The deep learning model achieves high macro-averaging scores for precision, recall, and F1-measure. These findings demonstrate the effectiveness of machine learning and deep learning methods in optimizing the detection model and improving glucose prediction accuracy. Theoretical Importance: This research contributes to the field by providing a comprehensive analysis of various machine-learning techniques for estimating glucose concentrations from NIR spectra. It also explores the use of deep learning for the classification of indistinguishable NIR spectra. The findings highlight the potential of machine learning and deep learning in enhancing the prediction accuracy of glucose-relevant features. Data Collection and Analysis Procedures: The NIR spectra and corresponding references for glucose concentration are measured in increments of 20 mg/dl. The data is randomly divided into train and test sets, and the models are evaluated using regression analysis and classification metrics. The performance of each model is assessed based on correlation coefficients, determination coefficients, precision, recall, and F1-measure. Question Addressed: The study addresses the question of whether machine learning and deep learning methods can optimize the detection model and improve the accuracy of glucose prediction from NIR spectra. Conclusion: The research demonstrates that machine learning and deep learning methods can effectively predict glucose concentration from NIR spectra. The SVMR, ETR, and PCA-NN models exhibit superior performance, while the deep learning model achieves high classification scores. These findings suggest that machine learning and deep learning techniques can be used to improve the prediction accuracy of glucose-relevant features. Further research is needed to explore their clinical utility in analyzing complex matrices, such as blood glucose levels.

Keywords: machine learning, signal processing, near-infrared spectroscopy, support vector machine, neural network

Procedia PDF Downloads 54

Authors: Jorge Sandoval Lezama, Arturo Ivan Sandoval Rodriguez, Jose Arturo Correa Arredondo

Abstract:

This work aims to share innovative educational experiences at IPN Mexico, that involve collaborative learning at institutional and global level through course competition and global collaboration projects. Students from universities in China, USA, South Korea, Canada and Mexico collaborate to design electric vehicles to solve global urban mobility problems. The participation of IPN students in the 2015-2016 global competition (São Paolo, Brazil and Cincinnati, USA) Reconfigurable Shared-Use Mobility Systems allowed to apply pedagogical strategies of groups of collaboration and of learning based on projects where they shared activities, commitments and goals, demonstrating that students were motivated to develop / self-generate their knowledge with greater meaning and understanding. One of the most evident achievements is that the students are self-managed, so the most advanced students train the students who join the project with CAD, CAE, CAM tools. Likewise, the motivation achieved is evident since in 2014 there were 12 students involved in the project, and there are currently more than 70 students.

Keywords: collaboration projects, global competency, course competition, active learning

Procedia PDF Downloads 239

Authors: Ashkan Zakaryazad, Ekrem Duman

Abstract:

A typical classification technique ranks the instances in a data set according to the likelihood of belonging to one (positive) class. A credit card (CC) fraud detection model ranks the transactions in terms of probability of being fraud. In fact, this approach is often criticized, because firms do not care about fraud probability but about the profitability or costliness of detecting a fraudulent transaction. The key contribution in this study is to focus on the profit maximization in the model building step. The artificial neural network proposed in this study works based on profit maximization instead of minimizing the error of prediction. Moreover, some studies have shown that the back propagation algorithm, similar to other gradient–based algorithms, usually gets trapped in local optima and swarm-based algorithms are more successful in this respect. In this study, we train our profit maximization ANN using the Migrating Birds optimization (MBO) which is introduced to literature recently.

Keywords: neural network, profit-based neural network, sum of squared errors (SSE), MBO, gradient descent

Procedia PDF Downloads 440

Authors: David F. Nettleton, Elodie Bugnicourt, Christian Wasiak, Alejandro Rosales

Abstract:

In this presentation, preliminary results are given for the modeling and calibration of two different industrial winding MIMO (Multiple Input Multiple Output) processes using machine learning techniques. In contrast to previous approaches which have typically used ‘black-box’ linear statistical methods together with a definition of the mechanical behavior of the process, we use non-linear machine learning algorithms together with a ‘white-box’ rule induction technique to create a supervised model of the fitting error between the expected and real force measures. The final objective is to build a precise model of the winding process in order to control de-tension of the material being wound in the first case, and the friction of the material passing through the die, in the second case. Case 1, Tension Control of a Winding Process. A plastic web is unwound from a first reel, goes over a traction reel and is rewound on a third reel. The objectives are: (i) to train a model to predict the web tension and (ii) calibration to find the input values which result in a given tension. Case 2, Friction Force Control of a Micro-Pullwinding Process. A core+resin passes through a first die, then two winding units wind an outer layer around the core, and a final pass through a second die. The objectives are: (i) to train a model to predict the friction on die2; (ii) calibration to find the input values which result in a given friction on die2. Different machine learning approaches are tested to build models, Kernel Ridge Regression, Support Vector Regression (with a Radial Basis Function Kernel) and MPART (Rule Induction with continuous value as output). As a previous step, the MPART rule induction algorithm was used to build an explicative model of the error (the difference between expected and real friction on die2). The modeling of the error behavior using explicative rules is used to help improve the overall process model. Once the models are built, the inputs are calibrated by generating Gaussian random numbers for each input (taking into account its mean and standard deviation) and comparing the output to a target (desired) output until a closest fit is found. The results of empirical testing show that a high precision is obtained for the trained models and for the calibration process. The learning step is the slowest part of the process (max. 5 minutes for this data), but this can be done offline just once. The calibration step is much faster and in under one minute obtained a precision error of less than 1x10-3 for both outputs. To summarize, in the present work two processes have been modeled and calibrated. A fast processing time and high precision has been achieved, which can be further improved by using heuristics to guide the Gaussian calibration. Error behavior has been modeled to help improve the overall process understanding. This has relevance for the quick optimal set up of many different industrial processes which use a pull-winding type process to manufacture fibre reinforced plastic parts. Acknowledgements to the Openmind project which is funded by Horizon 2020 European Union funding for Research & Innovation, Grant Agreement number 680820

Keywords: data model, machine learning, industrial winding, calibration

Procedia PDF Downloads 211

Authors: Vijaya Yuvaram Singh V M, Kameshwar Rao J V

Abstract:

The challenge with development of neural network based methods for medical is the availability of data. Anatomical landmark detection in the medical domain is a process to find points on the x-ray scan report of the patient. Most of the time this task is done manually by trained professionals as it requires precision and domain knowledge. Traditionally object detection based methods are used for landmark detection. Here, we utilize reinforcement learning and query based method to train a single agent capable of detecting multiple landmarks. A deep Q network agent is trained to detect single and multiple landmarks present on hip and shoulder from x-ray scan of a patient. Here a single agent is trained to find multiple landmark making it superior to having individual agents per landmark. For the initial study, five images of different patients are used as the environment and tested the agents performance on two unseen images.

Keywords: reinforcement learning, medical landmark detection, multi target detection, deep neural network

Procedia PDF Downloads 115

Authors: Sean Paulsen, Michael Casey

Abstract:

In this work we present a self-supervised pretraining framework for transformers on functional Magnetic Resonance Imaging (fMRI) data. First, we pretrain our architecture on two self-supervised tasks simultaneously to teach the model a general understanding of the temporal and spatial dynamics of human auditory cortex during music listening. Our pretraining results are the first to suggest a synergistic effect of multitask training on fMRI data. Second, we finetune the pretrained models and train additional fresh models on a supervised fMRI classification task. We observe significantly improved accuracy on held-out runs with the finetuned models, which demonstrates the ability of our pretraining tasks to facilitate transfer learning. This work contributes to the growing body of literature on transformer architectures for pretraining and transfer learning with fMRI data, and serves as a proof of concept for our pretraining tasks and multitask pretraining on fMRI data.

Keywords: transfer learning, fMRI, self-supervised, brain decoding, transformer, multitask training

Procedia PDF Downloads 58

Authors: Josselyn Mata Calidonio, Arianna I. Maddox, Kimberly Hamad-Schifferli

Abstract:

Rapid diagnostics are critical infectious disease tools that are designed to detect a known biomarker using antibodies specific to that biomarker. However, a way to detect unknown viruses has not yet been achieved in a paper test format. We describe here a route to make an adaptable paper immunoassay that can detect an unknown biomarker, demonstrating it on SARS-CoV-2 variants. The immunoassay repurposes cross-reactive antibodies raised against the alpha variant. Gold nanoparticles of two different colors conjugated to two different antibodies create a colorimetric signal, and machine learning of the resulting colorimetric pattern is used to train the assay to discriminate between variants of alpha and Omicron BA.5. By using principal component analysis, the colorimetric test patterns can pick up and discriminate an unknown that it has not encountered before, Omicron BA.1. The test has an accuracy of 100% and a potential calculated discriminatory power of 900. We show that it can be used adaptively and that it can be used to pick up emerging variants without the need to raise new antibodies.

Keywords: adaptive immunoassay, detecting unknown viruses, gold nanoparticles, paper immunoassay, repurposing antibodies

Procedia PDF Downloads 79

Authors: Sukhbir Singh

Abstract:

This paper describes the modeling of novel solar air collector (NSAC) system by using artificial neural network (ANN) model. The objective of the study is to demonstrate the application of the ANN model to predict the performance of the NSAC with acetamide as a phase change material (PCM) storage. Input data set consist of time, solar intensity and ambient temperature wherever as outlet air temperature of NSAC was considered as output. Experiments were conducted between 9.00 and 24.00 h in June and July 2014 underneath the prevailing atmospheric condition of Kurukshetra (city of the India). After that, experimental results were utilized to train the back propagation neural network (BPNN) to predict the outlet air temperature of NSAC. The results of proposed algorithm show that the BPNN is effective tool for the prediction of responses. The BPNN predicted results are 99% in agreement with the experimental results.

Keywords: Evacuated tube solar air collector, Artificial neural network, Phase change material, solar air collector

Procedia PDF Downloads 87

Authors: Sanat Ramesh, Tarun Dutt, Abhilasha Aswal, Anushka Chandrababu, G. N. Srinivasa Prasanna

Abstract:

Optimal Schedule Generation for a large national railway network operating thousands of passenger trains with tens of thousands of kilometers of track is a grand computational challenge in itself. We present heuristics based on a Mixed Integer Program (MIP) formulation for local optimization. These methods provide flexibility in scheduling new trains with varying speed and delays and improve utilization of infrastructure. We propose methods that provide a robust solution with hundreds of trains being scheduled over a portion of the railway network without significant increases in delay. We also provide techniques to validate the nominal schedules thus generated over global correlated variations in travel times thereby enabling us to detect conflicts arising due to delays. Our validation results which assume only the support of the arrival and departure time distributions takes an order of few minutes for a portion of the network and is computationally efficient to handle the entire network.

Keywords: mixed integer programming, optimization, railway network, train scheduling

Procedia PDF Downloads 124

Authors: Shyam Mehraaj, Samanthi E. R. Siriwardana, Shehara A. K. G. H., Wanigasinghe N. T., Wandana R. A. K., Wedage C. V.

Abstract:

A presentation is a critical tool for conveying not just spoken information but also a wide spectrum of human emotions. The single most effective thing to make the presentation successful is to practice it beforehand. Preparing for a presentation has been shown to be essential for improving emotional control, intonation and prosody, pronunciation, and vocabulary, as well as the quality of the presentation slides. As a result, practicing has become one of the most critical parts of giving a good presentation. In this research, the main focus is to analyze the audio, video, and slides of the presentation uploaded by the presenters. This proposed solution is based on the Natural Language Processing and Computer Vision techniques to cater to the requirement for the presenter to do a presentation beforehand using a mobile responsive web application. The proposed system will assist in practicing the presentation beforehand by identifying the presenters’ emotions, body language, tonality, prosody, pronunciations and vocabulary, and presentation slides quality. Overall, the system will give a rating and feedback to the presenter about the performance so that the presenters’ can improve their presentation skills.

Keywords: presentation, self-evaluation, natural learning processing, computer vision

Procedia PDF Downloads 68

Authors: Yihao Kuang, Bowen Ding

Abstract:

With the increasing scale and complexity of knowledge graph, modern knowledge graph contains more and more types of entity, relationship, and attribute information. Therefore, in recent years, it has been a trend for knowledge graph inference to use reinforcement learning to deal with large-scale, incomplete, and noisy knowledge graph and improve the inference effect and interpretability. The Proximal Policy Optimization (PPO) algorithm utilizes a near-end strategy optimization approach. This allows for more extensive updates of policy parameters while constraining the update extent to maintain training stability. This characteristic enables PPOs to converge to improve strategies more rapidly, often demonstrating enhanced performance early in the training process. Furthermore, PPO has the advantage of offline learning, effectively utilizing historical experience data for training and enhancing sample utilization. This means that even with limited resources, PPOs can efficiently train for reinforcement learning tasks. Based on these characteristics, this paper aims to obtain better and more efficient inference effect by introducing PPO into knowledge inference technology.

Keywords: reinforcement learning, PPO, knowledge inference, supervised learning

Procedia PDF Downloads 31

Authors: Seong Hyeok Lee, Jin Wook Lee, Bu Seog Ju, Woo Young Jung

Abstract:

Korea Train eXpress (KTX) is now being operated, which allows Korea being one of the countries that operates the high-speed rail system. The high-speed rail has its advantage of short time transportation of population and materials, which lead to many researches performed in this matter. In the case of high speed classical trackbed system, the maintenance and usability of gravel ballast system is costly. Recently, the concrete trackbed structure has been introduced as a replacement of classical trackbed system. In this case, the sleeper plays a critical role. Current study investigated to develop the track sleepers readily applicable to the top of the asphalt trackbed, as part of the trcakbed study utilizing the asphalt material. Among many possible shapes and design of sleepers, current study proposed two types of wide-sleepers according to the shear-key installation method. The structural behavior analysis and safety evaluation on each case was conducted using Korean design standard.

Keywords: wide sleepers, asphalt, high-speed railway, shear-key

Procedia PDF Downloads 405

Authors: Kateryna Pilyarchuk

Abstract:

A dress has, since times immemorial, been used to communicate the wearer’s identity. When a new trend is born, fashionistas buy it not only with the purpose to beautify themselves, but also to acquire the collective identity. Fashion has become a means of narrating one’s stance and status. Thus, when one spends money on a brand, one pays for some unmaterial components associated with it. This paper will present some ways in which fashion magazines promote consumerism by drawing on women’s craving for collective identity and need to fill in their identity gap by means of a purchase. By applying the method of critical discursive psychology, it will present layers of ideology and positions that become visible in framing of the message in U.S. Harper’s Bazaar. In this context, fashion decisions that are presented to its readers will be critically evaluated from the gender perspective. It will be demonstrated that what is presented as a postfeminist choice in the neoliberal society is still, to a considerable extent, oppressive and driven by the male gaze. As the findings show, the contemporary female identities in fashion are still built on the principles of traditional femininity. Magazines and fashion discourse train women that they should fear being left out of fashion and, by extension, out of the category of the sexually appealing (from the male perspective).

Keywords: collective identity, critical discursive psychology, fashion discourse, identity gap

Procedia PDF Downloads 118

Authors: Michael Okeke, Andrew Blyth

Abstract:

Supervisory Control And Data Acquisition (SCADA) as the state of the art Industrial Control Systems (ICS) are used in many different critical infrastructures, from smart home to energy systems and from locomotives train system to planes. Security of SCADA systems is vital since many lives depend on it for daily activities and deviation from normal operation could be disastrous to the environment as well as lives. This paper describes how No-Trust-Zone (NTZ) architecture could be incorporated into SCADA Systems in order to reduce the chances of malicious intent. The architecture is made up of two distinctive parts which are; the field devices such as; sensors, PLCs pumps, and actuators. The second part of the architecture is designed following lambda architecture, which is made up of a detection algorithm based on Particle Swarm Optimization (PSO) and Hadoop framework for data processing and storage. Apache Spark will be a part of the lambda architecture for real-time analysis of packets for anomalies detection.

Keywords: industrial control system (ics, no-trust-zone (ntz), particle swarm optimisation (pso), supervisory control and data acquisition (scada), swarm intelligence (SI)

Procedia PDF Downloads 311

Authors: Achim Kampker, Heiner Hans Heimes, Mathias Ordung, Nemanja Sarovic

Abstract:

A key component of the drive train of electric vehicles is the lithium-ion battery system. Among various other components, such as the battery management system or the thermal management system, the battery system mostly consists of several cells which are integrated mechanically as well as electrically. Due to different vehicle concepts with regards to space, energy and power specifications, there is a variety of different battery systems. The corresponding assembly lines are specially designed for each battery concept. Minor changes to certain characteristics of the battery have a disproportionally high effect on the set-up effort in the form of high change-related costs. This paper will focus on battery systems which are made out of battery cells with a prismatic format. The product architecture and the assembly process will be analyzed in detail based on battery concepts of existing electric cars and key variety-causing drivers will be identified. On this basis, several measures will be presented and discussed on how to change the product architecture and the assembly process in order to reduce change-related costs.

Keywords: assembly, automotive industry, battery system, battery concept

Procedia PDF Downloads 271

Authors: Asra Yousuf, Ajaykumar Tannirkulum

Abstract:

Inferring rural-urban migration trends can help design effective policies that promote better urban planning and rural development. In this paper, we describe how machine learning algorithms can be applied to predict internal migration decisions of people. We consider data collected from household surveys in Tamil Nadu to train our model. To measure the performance of the model, we use data on past migration from National Sample Survey Organisation of India. The factors for training the model include socioeconomic characteristic of each individual like age, gender, place of residence, outstanding loans, strength of the household, etc. and his past migration history. We perform a comparative analysis of the performance of a number of machine learning algorithm to determine their prediction accuracy. Our results show that machine learning algorithms provide a stronger prediction accuracy as compared to statistical models. Our goal through this research is to propose the use of data science techniques in understanding human decisions and behaviour in developing countries.

Keywords: development, migration, internal migration, machine learning, prediction

Procedia PDF Downloads 241

Authors: Umair Rashid

Abstract:

Normally softmax loss is used as the supervision signal in face recognition (FR) system, and it boosts the separability of features. In the last two years, a number of techniques have been proposed by reformulating the original softmax loss to enhance the discriminating power of Deep Convolutional Neural Networks (DCNNs) for FR system. To learn angularly discriminative features Cosine-Margin based softmax has been adjusted as monotonically decreasing angular function, that is the main challenge for angular based softmax. On that issue, we propose monotonically decreasing element for Cosine-Margin based softmax and also, we discussed the effect of different monotonically decreasing parameters on angular Margin softmax for FR system. We train the model on publicly available dataset CASIA- WebFace via our proposed monotonically decreasing parameters for cosine function and the tests on YouTube Faces (YTF, Labeled Face in the Wild (LFW), VGGFace1 and VGGFace2 attain the state-of-the-art performance.

Keywords: deep convolutional neural networks, cosine margin face recognition, softmax loss, monotonically decreasing parameter

Procedia PDF Downloads 56