Search results for: Siamese neural network

Authors: A. Shoiynbek, K. Kozhakhmet, P. Menezes, D. Kuanyshbay, D. Bayazitov

Abstract:

Speech emotion recognition has received increasing research interest all through current years. There was used emotional speech that was collected under controlled conditions in most research work. Actors imitating and artificially producing emotions in front of a microphone noted those records. There are four issues related to that approach, namely, (1) emotions are not natural, and it means that machines are learning to recognize fake emotions. (2) Emotions are very limited by quantity and poor in their variety of speaking. (3) There is language dependency on SER. (4) Consequently, each time when researchers want to start work with SER, they need to find a good emotional database on their language. In this paper, we propose the approach to create an automatic tool for speech emotion extraction based on facial emotion recognition and describe the sequence of actions of the proposed approach. One of the first objectives of the sequence of actions is a speech detection issue. The paper gives a detailed description of the speech detection model based on a fully connected deep neural network for Kazakh and Russian languages. Despite the high results in speech detection for Kazakh and Russian, the described process is suitable for any language. To illustrate the working capacity of the developed model, we have performed an analysis of speech detection and extraction from real tasks.

Keywords: deep neural networks, speech detection, speech emotion recognition, Mel-frequency cepstrum coefficients, collecting speech emotion corpus, collecting speech emotion dataset, Kazakh speech dataset

Procedia PDF Downloads 101

Authors: Kamel Laidi, Khelifa Benmansour, Ouahid Bouchhida

Abstract:

Neural networks-based approach for 2-cells serial converter has been developed and implemented. The approach is based on a behavioural description of the different operating modes of the converter. Each operating mode represents a well-defined configuration, and for which is matched an operating zone satisfying given invariance conditions, depending on the capacitors' voltages and the load current of the converter. For each mode, a control vector whose components are the control signals to be applied to the converter switches has been associated. Therefore, the problem is reduced to a classification task of the different operating modes of the converter. The artificial neural nets-based approach, which constitutes a powerful tool for this kind of task, has been adopted and implemented. The application to a 2-cells chopper has allowed ensuring efficient and robust control of the load current and a high capacitors voltages balancing.

Keywords: neural nets, control, multicellular converters, 2-cells chopper

Procedia PDF Downloads 834

Authors: S. Mehrab Amiri, Nasser Talebbeydokhti

Abstract:

Modeling sediment transport processes by means of numerical approach often poses severe challenges. In this way, a number of techniques have been suggested to solve flow and sediment equations in decoupled, semi-coupled or fully coupled forms. Furthermore, in order to capture flow discontinuities, a number of techniques, like artificial viscosity and shock fitting, have been proposed for solving these equations which are mostly required careful calibration processes. In this research, a numerical scheme for solving shallow water and Exner equations in fully coupled form is presented. First-Order Centered scheme is applied for producing required numerical fluxes and the reconstruction process is carried out toward using Monotonic Upstream Scheme for Conservation Laws to achieve a high order scheme.  In order to satisfy C-property of the scheme in presence of bed topography, Surface Gradient Method is proposed. Combining the presented scheme with fourth order Runge-Kutta algorithm for time integration yields a competent numerical scheme. In addition, to handle non-prismatic channels problems, Cartesian Cut Cell Method is employed. A trained Multi-Layer Perceptron Artificial Neural Network which is of Feed Forward Back Propagation (FFBP) type estimates sediment flow discharge in the model rather than usual empirical formulas. Hydrodynamic part of the model is tested for showing its capability in simulation of flow discontinuities, transcritical flows, wetting/drying conditions and non-prismatic channel flows. In this end, dam-break flow onto a locally non-prismatic converging-diverging channel with initially dry bed conditions is modeled. The morphodynamic part of the model is verified simulating dam break on a dry movable bed and bed level variations in an alluvial junction. The results show that the model is capable in capturing the flow discontinuities, solving wetting/drying problems even in non-prismatic channels and presenting proper results for movable bed situations. It can also be deducted that applying Artificial Neural Network, instead of common empirical formulas for estimating sediment flow discharge, leads to more accurate results.

Keywords: artificial neural network, morphodynamic model, sediment continuity equation, shallow water equations

Procedia PDF Downloads 187

Authors: Usama Shahid

Abstract:

Be it written documents, news columns, or students' essays, verifying the content can be a time-consuming task. Apart from the spelling and grammar mistakes, the proofreader is also supposed to verify whether the content included in the essay or document is relevant or not. The irrelevant content in any document or essay is referred to as off-topic text and in this paper, we will address the problem of off-topic text detection from a document using machine learning techniques. Our study aims to identify the off-topic content from a document using Echo state network model and we will also compare data with other models. The previous study uses Convolutional Neural Networks and TFIDF to detect off-topic text. We will rearrange the existing datasets and take new classifiers along with new word embeddings and implement them on existing and new datasets in order to compare the results with the previously existing CNN model.

Keywords: off topic, text detection, eco state network, machine learning

Procedia PDF Downloads 85

Authors: Shubham Gogri, Lucia Florescu

Abstract:

Medical diagnosis and radiotherapy treatment planning using Computed Tomography (CT) rely on the quantitative accuracy and quality of the CT images. At the same time, requirements for CT imaging include reducing the radiation dose exposure to patients and minimizing scanning time. A solution to this is the sparse-view CT technique, based on a reduced number of projection views. This, however, introduces a new problem— the incomplete projection data results in lower quality of the reconstructed images. To tackle this issue, deep learning methods have been applied to enhance the quality of the sparse-view CT images. A first approach involved employing Mir-Net, a dedicated deep neural network designed for image enhancement. This showed promise, utilizing an intricate architecture comprising encoder and decoder networks, along with the incorporation of the Charbonnier Loss. However, this approach was computationally demanding. Subsequently, a specialized Generative Adversarial Network (GAN) architecture, rooted in the Pix2Pix framework, was implemented. This GAN framework involves a U-Net-based Generator and a Discriminator based on Convolutional Neural Networks. To bolster the GAN's performance, both Charbonnier and Wasserstein loss functions were introduced, collectively focusing on capturing minute details while ensuring training stability. The integration of the perceptual loss, calculated based on feature vectors extracted from the VGG16 network pretrained on the ImageNet dataset, further enhanced the network's ability to synthesize relevant images. A series of comprehensive experiments with clinical CT data were conducted, exploring various GAN loss functions, including Wasserstein, Charbonnier, and perceptual loss. The outcomes demonstrated significant image quality improvements, confirmed through pertinent metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) between the corrected images and the ground truth. Furthermore, learning curves and qualitative comparisons added evidence of the enhanced image quality and the network's increased stability, while preserving pixel value intensity. The experiments underscored the potential of deep learning frameworks in enhancing the visual interpretation of CT scans, achieving outcomes with SSIM values close to one and PSNR values reaching up to 76.

Keywords: generative adversarial networks, sparse view computed tomography, CT image correction, Mir-Net

Procedia PDF Downloads 161

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 369

Authors: Kudzanayi Chiteka, Wellington Makondo

Abstract:

The whole world is on the drive to go green owing to the negative effects of burning fossil fuels. Therefore, there is immediate need to identify and utilise alternative renewable energy sources. Among these energy sources solar energy is one of the most dominant in Zimbabwe. Solar power plants used to generate electricity are entirely dependent on solar radiation. For planning purposes, solar radiation values should be known in advance to make necessary arrangements to minimise the negative effects of the absence of solar radiation due to cloud cover and other naturally occurring phenomena. This research focused on the prediction of Global Horizontal Irradiation values for the sixth day given values for the past five days. Artificial intelligence techniques were used in this research. Three models were developed based on Support Vector Machines, Radial Basis Function, and Feed Forward Back-Propagation Artificial neural network. Results revealed that Support Vector Machines gives the best results compared to the other two with a mean absolute percentage error (MAPE) of 2%, Mean Absolute Error (MAE) of 0.05kWh/m²/day root mean square (RMS) error of 0.15kWh/m²/day and a coefficient of determination of 0.990. The other predictive models had prediction accuracies of MAPEs of 4.5% and 6% respectively for Radial Basis Function and Feed Forward Back-propagation Artificial neural network. These two models also had coefficients of determination of 0.975 and 0.970 respectively. It was found that prediction of GHI values for the future days is possible using artificial intelligence-based predictive models.

Keywords: solar energy, global horizontal irradiation, artificial intelligence, predictive models

Procedia PDF Downloads 273

Authors: D. Pugazhenthi, B. Sree Vidya

Abstract:

Cloud computing is one of the emerging technologies that enables end users to use the services of cloud on ‘pay per usage’ strategy. This technology grows in a fast pace and so is its security threat. One among the various services provided by cloud is storage. In this service, security plays a vital factor for both authenticating legitimate users and protection of information. This paper brings in efficient ways of authenticating users as well as securing information on the cloud. Initial phase proposed in this paper deals with an authentication technique using multi-factor and multi-dimensional authentication system with multi-level security. Unique identification and slow intrusive formulates an advanced reliability on user-behaviour based biometrics than conventional means of password authentication. By biometric systems, the accounts are accessed only by a legitimate user and not by a nonentity. The biometric templates employed here do not include single trait but multiple, viz., iris and finger prints. The coordinating stage of the authentication system functions on Ensemble Support Vector Machine (SVM) and optimization by assembling weights of base SVMs for SVM ensemble after individual SVM of ensemble is trained by the Artificial Fish Swarm Algorithm (AFSA). Thus it helps in generating a user-specific secure cryptographic key of the multimodal biometric template by fusion process. Data security problem is averted and enhanced security architecture is proposed using encryption and decryption system with double key cryptography based on Fuzzy Neural Network (FNN) for data storing and retrieval in cloud computing . The proposing scheme aims to protect the records from hackers by arresting the breaking of cipher text to original text. This improves the authentication performance that the proposed double cryptographic key scheme is capable of providing better user authentication and better security which distinguish between the genuine and fake users. Thus, there are three important modules in this proposed work such as 1) Feature extraction, 2) Multimodal biometric template generation and 3) Cryptographic key generation. The extraction of the feature and texture properties from the respective fingerprint and iris images has been done initially. Finally, with the help of fuzzy neural network and symmetric cryptography algorithm, the technique of double key encryption technique has been developed. As the proposed approach is based on neural networks, it has the advantage of not being decrypted by the hacker even though the data were hacked already. The results prove that authentication process is optimal and stored information is secured.

Keywords: artificial fish swarm algorithm (AFSA), biometric authentication, decryption, encryption, fingerprint, fusion, fuzzy neural network (FNN), iris, multi-modal, support vector machine classification

Procedia PDF Downloads 259

Authors: Saed Khawaldeh, Mohamed Elsharnouby, Alaa Eddin Alchalabi, Usama Pervaiz, Tajwar Aleef, Vu Hoang Minh

Abstract:

Taxonomic classification has a wide-range of applications such as finding out more about the evolutionary history of organisms that can be done by making a comparison between species living now and species that lived in the past. This comparison can be made using different kinds of extracted species’ data which include DNA sequences. Compared to the estimated number of the organisms that nature harbours, humanity does not have a thorough comprehension of which specific species they all belong to, in spite of the significant development of science and scientific knowledge over many years. One of the methods that can be applied to extract information out of the study of organisms in this regard is to use the DNA sequence of a living organism as a marker, thus making it available to classify it into a taxonomy. The classification of living organisms can be done in many machine learning techniques including Neural Networks (NNs). In this study, DNA sequences classification is performed using Convolutional Neural Networks (CNNs) which is a special type of NNs.

Keywords: deep networks, convolutional neural networks, taxonomic classification, DNA sequences classification

Procedia PDF Downloads 442

Authors: M. Riazat, H. Abdolvand, M. Baniassadi

Abstract:

In this present work, 3D reconstruction of cathode of SOFC is developed with various volume fractions and porosity. Three Phase Boundary (TPB) of construction of such derived micro structures is calculated. The neural network is used to optimize the porosity and volume fraction of each phase to reach a structure with maximum TPB.

Keywords: fuel cell, solid oxide, TPB, 3D reconstruction

Procedia PDF Downloads 324

Authors: Zhiyong Shan, Preethi Santhanam, Vinod Namboodiri, Rajiv Bagai

Abstract:

In recent years, the emerging network worms and attacks have distributive characteristics, which can spread globally in a very short time. Security management crossing networks to co-defense network-wide attacks and improve the efficiency of security administration is urgently needed. We propose a hierarchical distributed network security management system (HD-NSMS), which can integrate security management across multiple networks. First, we describe the system in macrostructure and microstructure; then discuss three key problems when building HD-NSMS: device model, alert mechanism, and emergency response mechanism; lastly, we describe the implementation of HD-NSMS. The paper is valuable for implementing NSMS in that it derives from a practical network security management system (NSMS).

Keywords: network security management, device organization, emergency response, cross-network

Procedia PDF Downloads 168

Authors: Christian H. Sanabria-Montaña, Rodrigo Huerta-Quintanilla

Abstract:

A lattice network is a special type of network in which all nodes have the same number of links, and its boundary conditions are periodic. The most basic lattice network is the ring, a one-dimensional network with periodic border conditions. In contrast, the Cartesian product of d rings forms a d-dimensional lattice network. An analytical expression currently exists for the clustering coefficient in this type of network, but the theoretical value is valid only up to certain connectivity value; in other words, the analytical expression is incomplete. Here we obtain analytically the clustering coefficient expression in d-dimensional lattice networks for any link density. Our analytical results show that the clustering coefficient for a lattice network with density of links that tend to 1, leads to the value of the clustering coefficient of a fully connected network. We developed a model on criminology in which the generalized clustering coefficient expression is applied. The model states that delinquents learn the know-how of crime business by sharing knowledge, directly or indirectly, with their friends of the gang. This generalization shed light on the network properties, which is important to develop new models in different fields where network structure plays an important role in the system dynamic, such as criminology, evolutionary game theory, econophysics, among others.

Keywords: clustering coefficient, criminology, generalized, regular network d-dimensional

Procedia PDF Downloads 411

Authors: Yasaman Mohammadi

Abstract:

Emotion regulation is a complex process that allows individuals to manage and modulate their emotional responses in order to adaptively respond to environmental demands. As individuals age, emotion regulation abilities may decline, leading to an increased vulnerability to mood disorders and other negative health outcomes. Advances in neuroimaging techniques have greatly enhanced our understanding of the neural substrates underlying emotion regulation and age-related changes in these neural systems. Additionally, genetic research has identified several candidate genes that may influence age-related changes in emotion regulation. In this paper, we review recent findings from neuroimaging and genetic research on age-related changes in the neural substrates of emotion regulation, highlighting the mechanisms and consequences of these changes. We also discuss potential interventions, including cognitive and behavioral approaches, that may be effective in mitigating age-related declines in emotion regulation. We propose that a better understanding of the mechanisms underlying age-related changes in emotion regulation may lead to the development of more targeted interventions aimed at promoting healthy emotional functioning in older adults. Overall, this paper highlights the importance of studying age-related changes in emotion regulation and provides a roadmap for future research in this field.

Keywords: emotion regulation, aging, neural substrates, neuroimaging, emotional functioning, healthy aging

Procedia PDF Downloads 112

Authors: Christos Theoharatos, Dimitris Tsourounis, Spiros Oikonomou, Andreas Makedonas

Abstract:

This paper focuses on the problem of visually inspecting and recognizing the road conditions in front of moving vehicles, targeting automotive scenarios. The goal of road inspection is to identify whether the road is slippery or not, as well as to detect possible anomalies on the road surface like potholes or body bumps/humps. Our work is based on an artificial intelligence methodology for real-time monitoring of road conditions in autonomous driving scenarios, using state-of-the-art deep convolutional neural network (CNN) techniques. Initially, the road and ego lane are segmented within the field of view of the camera that is integrated into the front part of the vehicle. A novel classification CNN is utilized to identify among plain and slippery road textures (e.g., wet, snow, etc.). Simultaneously, a robust detection CNN identifies severe surface anomalies within the ego lane, such as potholes and speed bumps/humps, within a distance of 5 to 25 meters. The overall methodology is illustrated under the scope of an integrated application (or system), which can be integrated into complete Advanced Driver-Assistance Systems (ADAS) systems that provide a full range of functionalities. The outcome of the proposed techniques present state-of-the-art detection and classification results and real-time performance running on AI accelerator devices like Intel’s Myriad 2/X Vision Processing Unit (VPU).

Keywords: deep learning, convolutional neural networks, road condition classification, embedded systems

Procedia PDF Downloads 134

Authors: Dávid Kovács, Bálint Csanády, Dániel Boros, Iván Ivkovic, Lóránt Nagy, Dalma Tóth-Lakits, László Márkus, András Lukács

Abstract:

The modeling practice of financial instruments has seen significant change over the last decade due to the recognition of time-dependent and stochastically changing correlations among the market prices or the prices and market characteristics. To represent this phenomenon, the Stochastic Correlation Process (SCP) has come to the fore in the joint modeling of prices, offering a more nuanced description of their interdependence. This approach has allowed for the attainment of realistic tail dependencies, highlighting that prices tend to synchronize more during intense or volatile trading periods, resulting in stronger correlations. Evidence in statistical literature suggests that, similarly to the volatility, the SCP of certain stock prices follows rough paths, which can be described using fractional differential equations. However, estimating parameters for these equations often involves complex and computation-intensive algorithms, creating a necessity for alternative solutions. In this regard, the Fractional Ornstein-Uhlenbeck (fOU) process from the family of fractional processes offers a promising path. We can effectively describe the rough SCP by utilizing certain transformations of the fOU. We employed neural networks to understand the behavior of these processes. We had to develop a fast algorithm to generate a valid and suitably large sample from the appropriate process to train the network. With an extensive training set, the neural network can estimate the process parameters accurately and efficiently. Although the initial focus was the fOU, the resulting model displayed broader applicability, thus paving the way for further investigation of other processes in the realm of financial mathematics. The utility of SCP extends beyond its immediate application. It also serves as a springboard for a deeper exploration of fractional processes and for extending existing models that use ordinary Wiener processes to fractional scenarios. In essence, deploying both SCP and fractional processes in financial models provides new, more accurate ways to depict market dynamics.

Keywords: fractional Ornstein-Uhlenbeck process, fractional stochastic processes, Heston model, neural networks, stochastic correlation, stochastic differential equations, stochastic volatility

Procedia PDF Downloads 118

Authors: Ravneil Nand

Abstract:

Cooperative coevolution uses problem decomposition methods to solve a larger problem. The problem decomposition deals with breaking down the larger problem into a number of smaller sub-problems depending on their method. Different problem decomposition methods have their own strengths and limitations depending on the neural network used and application problem. In this paper we are introducing a new problem decomposition method known as Hidden-Neuron Level Decomposition (HNL). The HNL method is competing with established problem decomposition method in time series prediction. The results show that the proposed approach has improved the results in some benchmark data sets when compared to the standalone method and has competitive results when compared to methods from literature.

Keywords: cooperative coevaluation, feed forward network, problem decomposition, neuron, synapse

Procedia PDF Downloads 335

Authors: Leonard Tiong, Sonit Singh, Kevin Ho Shon, Sarah Lewis

Abstract:

Investigation into the use of artificial intelligence in radiology continues to develop at a rapid rate. During the coronavirus pandemic, the combination of an exponential increase in chest x-rays and unpredictable staff shortages resulted in a huge strain on the department's workload. There is a World Health Organisation estimate that two-thirds of the global population does not have access to diagnostic radiology. Therefore, there could be demand for a program that could detect acute changes in imaging compatible with infection to assist with screening. We generated a conventional neural network and tested its efficacy in recognizing changes compatible with coronavirus infection. Following ethics approval, a deidentified set of 77 normal and 77 abnormal chest x-rays in patients with confirmed coronavirus infection were used to generate an algorithm that could train, validate and then test itself. DICOM and PNG image formats were selected due to their lossless file format. The model was trained with 100 images (50 positive, 50 negative), validated against 28 samples (14 positive, 14 negative), and tested against 26 samples (13 positive, 13 negative). The initial training of the model involved training a conventional neural network in what constituted a normal study and changes on the x-rays compatible with coronavirus infection. The weightings were then modified, and the model was executed again. The training samples were in batch sizes of 8 and underwent 25 epochs of training. The results trended towards an 85.71% true positive/true negative detection rate and an area under the curve trending towards 0.95, indicating approximately 95% accuracy in detecting changes on chest X-rays compatible with coronavirus infection. Study limitations include access to only a small dataset and no specificity in the diagnosis. Following a discussion with our programmer, there are areas where modifications in the weighting of the algorithm can be made in order to improve the detection rates. Given the high detection rate of the program, and the potential ease of implementation, this would be effective in assisting staff that is not trained in radiology in detecting otherwise subtle changes that might not be appreciated on imaging. Limitations include the lack of a differential diagnosis and application of the appropriate clinical history, although this may be less of a problem in day-to-day clinical practice. It is nonetheless our belief that implementing this program and widening its scope to detecting multiple pathologies such as lung masses will greatly assist both the radiology department and our colleagues in increasing workflow and detection rate.

Keywords: artificial intelligence, COVID, neural network, machine learning

Procedia PDF Downloads 93

Authors: Sasiphan Nitayaprapha

Abstract:

A commercially-supported social network has become an emerging channel for an organization to communicate with and provide services to customers. The success of the commercially-supported social network depends on the ability of the organization to keep the customers in participating in the network. Drawing from the theories of information adoption, information systems continuance, and web usability, the author develops a model to explore how a commercially-supported social network can encourage customers to continue participating and using the information in the network. The theoretical model will be proved through an online survey of customers using the commercially-supported social networking sites of several high technology companies operating in the same sector. The result will be compared with previous studies to learn about the explanatory power of the research model, and to identify the main factors determining users’ intention to continue using a commercially-supported social network. Theoretical and practical implications, and limitations are discussed.

Keywords: social network, information adoption, information systems continuance, web usability, user satisfaction

Procedia PDF Downloads 316

Authors: Mecheri Zeid Belmecheri, Maamar Ahfir, Izzet Kale

Abstract:

Automatic cardiac auscultation is still a subject of research in order to establish an objective diagnosis. Recorded heart sounds as Phonocardiogram signals (PCG) can be used for automatic segmentation into components that have clinical meanings. These are the first sound, S1, the second sound, S2, and the systolic and diastolic components, respectively. In this paper, an automatic method is proposed for the robust segmentation of heart sounds. This method is based on calculating an intermediate sawtooth-shaped signal from the length variation of the recorded Phonocardiogram (PCG) signal in the time domain and, using its positive derivative function that is a binary signal in training a Recurrent Neural Network (RNN). Results obtained in the context of a large database of recorded PCGs with their simultaneously recorded ElectroCardioGrams (ECGs) from different patients in clinical settings, including normal and abnormal subjects, show a segmentation testing performance average of 76 % sensitivity and 94 % specificity.

Keywords: heart sounds, PCG segmentation, event detection, recurrent neural networks, PCG curve length

Procedia PDF Downloads 178

Authors: Xiaolei Hu, Enrico Ferrera, Riccardo Tomasi, Claudio Pastrone

Abstract:

Load Forecasting plays a key role in making today's and future's Smart Energy Grids sustainable and reliable. Accurate power consumption prediction allows utilities to organize in advance their resources or to execute Demand Response strategies more effectively, which enables several features such as higher sustainability, better quality of service, and affordable electricity tariffs. It is easy yet effective to apply Load Forecasting at larger geographic scale, i.e. Smart Micro Grids, wherein the lower available grid flexibility makes accurate prediction more critical in Demand Response applications. This paper analyses the application of short-term load forecasting in a concrete scenario, proposed within the EU-funded GreenCom project, which collect load data from single loads and households belonging to a Smart Micro Grid. Three short-term load forecasting techniques, i.e. linear regression, artificial neural networks, and radial basis function network, are considered, compared, and evaluated through absolute forecast errors and training time. The influence of weather conditions in Load Forecasting is also evaluated. A new definition of Gain is introduced in this paper, which innovatively serves as an indicator of short-term prediction capabilities of time spam consistency. Two models, 24- and 1-hour-ahead forecasting, are built to comprehensively compare these three techniques.

Keywords: short-term load forecasting, smart micro grid, linear regression, artificial neural networks, radial basis function network, gain

Procedia PDF Downloads 468

Authors: Mehrdad Shafiei Dizaji, Hoda Azari

Abstract:

The research explores the pioneering integration of Physics-Informed Neural Networks (PINNs) into the domain of Ground-Penetrating Radar (GPR) data prediction, akin to advancements in medical imaging for tracking tumor progression in the human body. This research presents a detailed development framework for a specialized PINN model proficient at interpreting and forecasting GPR data, much like how medical imaging models predict tumor behavior. By harnessing the synergy between deep learning algorithms and the physical laws governing subsurface structures—or, in medical terms, human tissues—the model effectively embeds the physics of electromagnetic wave propagation into its architecture. This ensures that predictions not only align with fundamental physical principles but also mirror the precision needed in medical diagnostics for detecting and monitoring tumors. The suggested deep learning structure comprises three components: a CNN, a spatial feature channel attention (SFCA) mechanism, and ConvLSTM, along with temporal feature frame attention (TFFA) modules. The attention mechanism computes channel attention and temporal attention weights using self-adaptation, thereby fine-tuning the visual and temporal feature responses to extract the most pertinent and significant visual and temporal features. By integrating physics directly into the neural network, our model has shown enhanced accuracy in forecasting GPR data. This improvement is vital for conducting effective assessments of bridge deck conditions and other evaluations related to civil infrastructure. The use of Physics-Informed Neural Networks (PINNs) has demonstrated the potential to transform the field of Non-Destructive Evaluation (NDE) by enhancing the precision of infrastructure deterioration predictions. Moreover, it offers a deeper insight into the fundamental mechanisms of deterioration, viewed through the prism of physics-based models.

Keywords: physics-informed neural networks, deep learning, ground-penetrating radar (GPR), NDE, ConvLSTM, physics, data driven

Procedia PDF Downloads 40

Authors: Yu-Jia Jian, Emily Chia-Yu Su, Hui-Ling Hsu, Jian-Jhih Chen

Abstract:

Coronary artery is the major supplier of myocardial blood flow. When fat and cholesterol are deposit in the coronary arterial wall, narrowing and stenosis of the artery occurs, which may lead to myocardial ischemia and eventually infarction. According to the World Health Organization (WHO), estimated 740 million people have died of coronary heart disease in 2015. According to Statistics from Ministry of Health and Welfare in Taiwan, heart disease (except for hypertensive diseases) ranked the second among the top 10 causes of death from 2013 to 2016, and it still shows a growing trend. According to American Heart Association (AHA), the risk factors for coronary heart disease including: age (> 65 years), sex (men to women with 2:1 ratio), obesity, diabetes, hypertension, hyperlipidemia, smoking, family history, lack of exercise and more. We have collected a dataset of 421 patients from a hospital located in northern Taiwan who received coronary computed tomography (CT) angiography. There were 300 males (71.26%) and 121 females (28.74%), with age ranging from 24 to 92 years, and a mean age of 56.3 years. Prior to coronary CT angiography, basic data of the patients, including age, gender, obesity index (BMI), diastolic blood pressure, systolic blood pressure, diabetes, hypertension, hyperlipidemia, smoking, family history of coronary heart disease and exercise habits, were collected and used as input variables. The output variable of the prediction module is the degree of coronary artery stenosis. The output variable of the prediction module is the narrow constriction of the coronary artery. In this study, the dataset was randomly divided into 80% as training set and 20% as test set. Four machine learning algorithms, including logistic regression, stepwise regression, neural network and decision tree, were incorporated to generate prediction results. We used area under curve (AUC) / accuracy (Acc.) to compare the four models, the best model is neural network, followed by stepwise logistic regression, decision tree, and logistic regression, with 0.68 / 79 %, 0.68 / 74%, 0.65 / 78%, and 0.65 / 74%, respectively. Sensitivity of neural network was 27.3%, specificity was 90.8%, stepwise Logistic regression sensitivity was 18.2%, specificity was 92.3%, decision tree sensitivity was 13.6%, specificity was 100%, logistic regression sensitivity was 27.3%, specificity 89.2%. From the result of this study, we hope to improve the accuracy by improving the module parameters or other methods in the future and we hope to solve the problem of low sensitivity by adjusting the imbalanced proportion of positive and negative data.

Keywords: decision support, computed tomography, coronary artery, machine learning

Procedia PDF Downloads 229

Authors: Ming-Jen Huang, Chun-Fang Huang, Chiching Wei

Abstract:

With the recent advance of the deep neural network, we observe new applications of NLP (natural language processing) and CV (computer vision) powered by deep neural networks for processing business documents. However, creating a real-world document processing system needs to integrate several NLP and CV tasks, rather than treating them separately. There is a need to have a unified approach for processing documents containing textual and graphical elements with rich formats, diverse layout arrangement, and distinct semantics. In this paper, a framework that fulfills this unified approach is presented. The framework includes a representation model definition for holding the information generated by various tasks and specifications defining the coordination between these tasks. The framework is a blueprint for building a system that can process documents with rich formats, styles, and multiple types of elements. The flexible and lightweight design of the framework can help build a system for diverse business scenarios, such as contract monitoring and reviewing.

Keywords: document processing, framework, formal definition, machine learning

Procedia PDF Downloads 214

Authors: O. A. Lawal, O. A Ojesanmi

Abstract:

The demand for wireless cellular services continues to increase while the radio resources remain limited. Thus, network operators have to continuously manage the scarce radio resources in order to have an improved quality of service for mobile users. This paper proposes how to handle the problem of congestion in the mobile network by prioritizing handoff call, using the guard channel allocation scheme. The research uses specific threshold value for the time of allocation of the channel in the algorithm. The scheme would be simulated by generating various data for different traffics in the network as it would be in the real life. The result would be used to determine the probability of handoff call dropping and the probability of the new call blocking as a way of measuring the network performance.

Keywords: call block, channel, handoff, mobile cellular network

Procedia PDF Downloads 394

Authors: Lei Zhu, Nan Li

Abstract:

Cold stamping has been widely applied in the automotive industry for the mass production of a great range of automotive panels. Predicting the springback to ensure the dimensional accuracy of the cold-stamped components is a critical step. The main approaches for the prediction and compensation of springback in cold stamping include running Finite Element (FE) simulations and conducting experiments, which require forming process expertise and can be time-consuming and expensive for the design of cold stamping tools. Machine learning technologies have been proven and successfully applied in learning complex system behaviours using presentative samples. These technologies exhibit the promising potential to be used as supporting design tools for metal forming technologies. This study, for the first time, presents a novel application of a Convolutional Neural Network (CNN) based surrogate model to predict the springback fields for variable U-shape cold bending geometries. A dataset is created based on the U-shape cold bending geometries and the corresponding FE simulations results. The dataset is then applied to train the CNN surrogate model. The result shows that the surrogate model can achieve near indistinguishable full-field predictions in real-time when compared with the FE simulation results. The application of CNN in efficient springback prediction can be adopted in industrial settings to aid both conceptual and final component designs for designers without having manufacturing knowledge.

Keywords: springback, cold stamping, convolutional neural networks, machine learning

Procedia PDF Downloads 149

Authors: Gong Zhilin, Jing Yang, Jian Yin

Abstract:

The purpose of this paper is to develop a novel system for credit card fraud detection based on sequential modeling of data using hybrid deep learning models. The projected model encapsulates five major phases are pre-processing, imbalance-data handling, feature extraction, optimal feature selection, and fraud detection with an ensemble classifier. The collected raw data (input) is pre-processed to enhance the quality of the data through alleviation of the missing data, noisy data as well as null values. The pre-processed data are class imbalanced in nature, and therefore they are handled effectively with the K-means clustering-based SMOTE model. From the balanced class data, the most relevant features like improved Principal Component Analysis (PCA), statistical features (mean, median, standard deviation) and higher-order statistical features (skewness and kurtosis). Among the extracted features, the most optimal features are selected with the Self-improved Arithmetic Optimization Algorithm (SI-AOA). This SI-AOA model is the conceptual improvement of the standard Arithmetic Optimization Algorithm. The deep learning models like Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and optimized Quantum Deep Neural Network (QDNN). The LSTM and CNN are trained with the extracted optimal features. The outcomes from LSTM and CNN will enter as input to optimized QDNN that provides the final detection outcome. Since the QDNN is the ultimate detector, its weight function is fine-tuned with the Self-improved Arithmetic Optimization Algorithm (SI-AOA).

Keywords: credit card, data mining, fraud detection, money transactions

Procedia PDF Downloads 131

Authors: Nazmul Shahadat, Anthony S. Maida

Abstract:

In recent years, hypercomplex inspired neural networks improved deep CNN architectures due to their ability to share weights across input channels and thus improve cohesiveness of representations within the layers. The work described herein studies the effect of replacing existing layers in an Axial Attention ResNet with their quaternion variants that use cross-channel weight sharing to assess the effect on image classification. We expect the quaternion enhancements to produce improved feature maps with more interlinked representations. We experiment with the stem of the network, the bottleneck layer, and the fully connected backend by replacing them with quaternion versions. These modifications lead to novel architectures which yield improved accuracy performance on the ImageNet300k classification dataset. Our baseline networks for comparison were the original real-valued ResNet, the original quaternion-valued ResNet, and the Axial Attention ResNet. Since improvement was observed regardless of which part of the network was modified, there is a promise that this technique may be generally useful in improving classification accuracy for a large class of networks.

Keywords: axial attention, representational networks, weight sharing, cross-channel correlations, quaternion-enhanced axial attention, deep networks

Procedia PDF Downloads 83

Authors: Kamalpreet Kaur, Harjit Pal Singh, Vikas Khullar

Abstract:

In Wireless Sensor Network (WSN), the main limitation is generally inimitable energy consumption during processing of the sensor nodes. Cluster head (CH) election is one of the main issues that can reduce the energy consumption. Therefore, discovering energy saving routing protocol is the focused area for research. In this paper, fuzzy-based energy aware routing protocol is presented, which enhances the stability and network lifetime of the network. Fuzzy logic ensures the well-organized selection of CH by taking four linguistic variables that are concentration, energy, centrality, and distance to base station (BS). The results show that the proposed protocol shows better results in requisites of stability and throughput of the network.

Keywords: ERA, fuzzy logic, network model, WSN

Procedia PDF Downloads 278

Authors: Khairani Binti Supyan, Fatimah Khalid, Mas Rina Mustaffa, Azreen Bin Azman, Amirul Azuani Romle

Abstract:

Perennial plant classification plays a significant role in various agricultural and environmental applications, assisting in plant identification, disease detection, and biodiversity monitoring. Nevertheless, attaining high accuracy in perennial plant image classification remains challenging due to the complex variations in plant appearance, the diverse range of environmental conditions under which images are captured, and the inherent variability in image quality stemming from various factors such as lighting conditions, camera settings, and focus. This paper proposes an adaptation approach to optimize perennial plant image classification by fine-tuning the pre-trained DNNs model. This paper explores the efficacy of fine-tuning prevalent architectures, namely VGG16, ResNet50, and InceptionV3, leveraging transfer learning to tailor the models to the specific characteristics of perennial plant datasets. A subset of the MYLPHerbs dataset consisted of 6 perennial plant species of 13481 images under various environmental conditions that were used in the experiments. Different strategies for fine-tuning, including adjusting learning rates, training set sizes, data augmentation, and architectural modifications, were investigated. The experimental outcomes underscore the effectiveness of fine-tuning deep neural networks for perennial plant image classification, with ResNet50 showcasing the highest accuracy of 99.78%. Despite ResNet50's superior performance, both VGG16 and InceptionV3 achieved commendable accuracy of 99.67% and 99.37%, respectively. The overall outcomes reaffirm the robustness of the fine-tuning approach across different deep neural network architectures, offering insights into strategies for optimizing model performance in the domain of perennial plant image classification.

Keywords: perennial plants, image classification, deep neural networks, fine-tuning, transfer learning, VGG16, ResNet50, InceptionV3

Procedia PDF Downloads 64

Authors: Alexander Hudson, Shaogang Gong

Abstract:

Structure determination is key to understanding protein function at a molecular level. Whilst significant advances have been made in predicting structure and function from amino acid sequence, researchers must still rely on expensive, time-consuming analytical methods to visualise detailed protein conformation. In this study, we demonstrate that it is possible to make accurate (≥80%) predictions of protein class and architecture from structures determined at low (>3A) resolution, using a deep convolutional neural network trained on high-resolution (≤3A) structures represented as 2D matrices. Thus, we provide proof of concept for high-speed, low-cost protein structure classification at low resolution, and a basis for extension to prediction of function. We investigate the impact of the input representation on classification performance, showing that side-chain information may not be necessary for fine-grained structure predictions. Finally, we confirm that high resolution, low-resolution and NMR-determined structures inhabit a common feature space, and thus provide a theoretical foundation for boosting with single-image super-resolution.

Keywords: transfer learning, protein distance maps, protein structure classification, neural networks

Procedia PDF Downloads 136