Search results for: neural network models

Authors: Farid Khosravikia, Patricia Clayton

Abstract:

The main objective of this study is to evaluate the advantages and disadvantages of various machine learning techniques in forecasting ground-motion intensity measures given source characteristics, source-to-site distance, and local site condition. Intensity measures such as peak ground acceleration and velocity (PGA and PGV, respectively) as well as 5% damped elastic pseudospectral accelerations at different periods (PSA), are indicators of the strength of shaking at the ground surface. Estimating these variables for future earthquake events is a key step in seismic hazard assessment and potentially subsequent risk assessment of different types of structures. Typically, linear regression-based models, with pre-defined equations and coefficients, are used in ground motion prediction. However, due to the restrictions of the linear regression methods, such models may not capture more complex nonlinear behaviors that exist in the data. Thus, this study comparatively investigates potential benefits from employing other machine learning techniques as a statistical method in ground motion prediction such as Artificial Neural Network, Random Forest, and Support Vector Machine. The algorithms are adjusted to quantify event-to-event and site-to-site variability of the ground motions by implementing them as random effects in the proposed models to reduce the aleatory uncertainty. All the algorithms are trained using a selected database of 4,528 ground-motions, including 376 seismic events with magnitude 3 to 5.8, recorded over the hypocentral distance range of 4 to 500 km in Oklahoma, Kansas, and Texas since 2005. The main reason of the considered database stems from the recent increase in the seismicity rate of these states attributed to petroleum production and wastewater disposal activities, which necessities further investigation in the ground motion models developed for these states. Accuracy of the models in predicting intensity measures, generalization capability of the models for future data, as well as usability of the models are discussed in the evaluation process. The results indicate the algorithms satisfy some physically sound characteristics such as magnitude scaling distance dependency without requiring pre-defined equations or coefficients. Moreover, it is shown that, when sufficient data is available, all the alternative algorithms tend to provide more accurate estimates compared to the conventional linear regression-based method, and particularly, Random Forest outperforms the other algorithms. However, the conventional method is a better tool when limited data is available.

Keywords: artificial neural network, ground-motion models, machine learning, random forest, support vector machine

Procedia PDF Downloads 123

Authors: Tesfaye Mengistu

Abstract:

Power cables play a crucial role in the transmission and distribution of electrical energy. As the electricity generation, transmission, distribution, and storage systems become smarter, there is a growing emphasis on incorporating intelligent approaches to ensure the reliability of power cables. Various types of electrical cables are employed for transmitting and distributing electrical energy, with cross-linked polyethylene (XLPE) cables being widely utilized due to their exceptional electrical and mechanical properties. However, insulation defects can occur in XLPE cables due to subpar manufacturing techniques during production and cable joint installation. To address this issue, experts have proposed different methods for monitoring XLPE cables. Some suggest the use of interdigital capacitive (IDC) technology for online monitoring, while others propose employing continuous wave (CW) terahertz (THz) imaging systems to detect internal defects in XLPE plates used for power cable insulation. In this study, we have developed models that employ a custom dataset collected locally to classify the physical safety status of individual power cables. Our models aim to replace physical inspections with computer vision and image processing techniques to classify defective power cables from non-defective ones. The implementation of our project utilized the Python programming language along with the TensorFlow package and a convolutional neural network (CNN). The CNN-based algorithm was specifically chosen for power cable defect classification. The results of our project demonstrate the effectiveness of CNNs in accurately classifying power cable defects. We recommend the utilization of similar or additional datasets to further enhance and refine our models. Additionally, we believe that our models could be used to develop methodologies for detecting power cable defects from live video feeds. We firmly believe that our work makes a significant contribution to the field of power cable inspection and maintenance. Our models offer a more efficient and cost-effective approach to detecting power cable defects, thereby improving the reliability and safety of power grids.

Keywords: artificial intelligence, computer vision, defect detection, convolutional neural net

Procedia PDF Downloads 114

Authors: Sinarwati Mohamad Suhaili, Naomie Salim, Mohamad Nazim Jambli

Abstract:

Sequence-to-sequence (seq2seq) models augmented with attention mechanisms are playing an increasingly important role in automated customer service. These models, which are able to recognize complex relationships between input and output sequences, are crucial for optimizing chatbot responses. Central to these mechanisms are neural attention weights that determine the focus of the model during sequence generation. Despite their widespread use, there remains a gap in the comparative analysis of different attention weighting functions within seq2seq models, particularly in the domain of chatbots using the Customer Support Twitter (CST) dataset. This study addresses this gap by evaluating four distinct attention-scoring functions—dot, multiplicative/general, additive, and an extended multiplicative function with a tanh activation parameter — in neural generative seq2seq models. Utilizing the CST dataset, these models were trained and evaluated over 10 epochs with the AdamW optimizer. Evaluation criteria included validation loss and BLEU scores implemented under both greedy and beam search strategies with a beam size of k=3. Results indicate that the model with the tanh-augmented multiplicative function significantly outperforms its counterparts, achieving the lowest validation loss (1.136484) and the highest BLEU scores (0.438926 under greedy search, 0.443000 under beam search, k=3). These results emphasize the crucial influence of selecting an appropriate attention-scoring function in improving the performance of seq2seq models for chatbots. Particularly, the model that integrates tanh activation proves to be a promising approach to improve the quality of chatbots in the customer support context.

Keywords: attention weight, chatbot, encoder-decoder, neural generative attention, score function, sequence-to-sequence

Procedia PDF Downloads 78

Authors: H. Bonakdari, I. Ebtehaj

Abstract:

The prediction of scour depth around bridge piers is frequently considered in river engineering. One of the key aspects in efficient and optimum bridge structure design is considered to be scour depth estimation around bridge piers. In this study, scour depth around bridge piers is estimated using two methods, namely the Adaptive Neuro-Fuzzy Inference System (ANFIS) and Artificial Neural Network (ANN). Therefore, the effective parameters in scour depth prediction are determined using the ANN and ANFIS methods via dimensional analysis, and subsequently, the parameters are predicted. In the current study, the methods’ performances are compared with the nonlinear regression (NLR) method. The results show that both methods presented in this study outperform existing methods. Moreover, using the ratio of pier length to flow depth, ratio of median diameter of particles to flow depth, ratio of pier width to flow depth, the Froude number and standard deviation of bed grain size parameters leads to optimal performance in scour depth estimation.

Keywords: adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), bridge pier, scour depth, nonlinear regression (NLR)

Procedia PDF Downloads 221

Authors: Kyung-Jin You, Kiwon Rhee, Marc H. Schieber, Nitish V. Thakor, Hyun-Chool Shin

Abstract:

It remains unknown whether M1 neurons encode multi-finger movements independently or as a certain neural network of single finger movements although multi-finger movements are physically a combination of single finger movements. We present an evidence of correlation between single and multi-finger movements and also attempt a challenging task of semi-blind decoding of neural data with minimum training of the neural decoder. Data were collected from 115 task-related neurons in M1 of a trained rhesus monkey performing flexion and extension of each finger and the wrist (12 single and 6 two-finger-movements). By exploiting correlation of temporal firing pattern between movements, we found that correlation coefficient for physically related movements pairs is greater than others; neurons tuned to single finger movements increased their firing rate when multi-finger commands were instructed. According to this knowledge, neural semi-blind decoding is done by choosing the greatest and the second greatest likelihood for canonical candidates. We achieved a decoding accuracy about 60% for multiple finger movement without corresponding training data set. this results suggest that only with the neural activities on single finger movements can be exploited to control dexterous multi-fingered neuroprosthetics.

Keywords: finger movement, neural activity, blind decoding, M1

Procedia PDF Downloads 323

Authors: Mo-Se Lee, Cheol-Hwi Ahn, Kee-Young Kwahk, Hyunchul Ahn

Abstract:

Over the past decade, deep learning has been in spotlight among various machine learning algorithms. In particular, CNN (Convolutional Neural Network), which is known as effective solution for recognizing and classifying images, has been popularly applied to classification and prediction problems in various fields. In this study, we try to apply CNN to stock market prediction, one of the most challenging tasks in the machine learning research. In specific, we propose to apply CNN as the binary classifier that predicts stock market direction (up or down) by using a graph as its input. That is, our proposal is to build a machine learning algorithm that mimics a person who looks at the graph and predicts whether the trend will go up or down. Our proposed model consists of four steps. In the first step, it divides the dataset into 5 days, 10 days, 15 days, and 20 days. And then, it creates graphs for each interval in step 2. In the next step, CNN classifiers are trained using the graphs generated in the previous step. In step 4, it optimizes the hyper parameters of the trained model by using the validation dataset. To validate our model, we will apply it to the prediction of KOSPI200 for 1,986 days in eight years (from 2009 to 2016). The experimental dataset will include 14 technical indicators such as CCI, Momentum, ROC and daily closing price of KOSPI200 of Korean stock market.

Keywords: convolutional neural network, deep learning, Korean stock market, stock market prediction

Procedia PDF Downloads 425

Authors: Jingting Luo, Yong Wang, Ying Wang

Abstract:

Pedestrian detection in image or video data is a very important and challenging task in security surveillance. The difficulty of this task is to locate and detect pedestrians of different scales in complex scenes accurately. To solve these problems, a deep neural network (RT-YOLOv3) is proposed to realize real-time pedestrian detection at different scales in security monitoring. RT-YOLOv3 improves the traditional YOLOv3 algorithm. Firstly, the deep residual network is added to extract vehicle features. Then six convolutional neural networks with different scales are designed and fused with the corresponding scale feature maps in the residual network to form the final feature pyramid to perform pedestrian detection tasks. This method can better characterize pedestrians. In order to further improve the accuracy and generalization ability of the model, a hybrid pedestrian data set training method is used to extract pedestrian data from the VOC data set and train with the INRIA pedestrian data set. Experiments show that the proposed RT-YOLOv3 method achieves 93.57% accuracy of mAP (mean average precision) and 46.52f/s (number of frames per second). In terms of accuracy, RT-YOLOv3 performs better than Fast R-CNN, Faster R-CNN, YOLO, SSD, YOLOv2, and YOLOv3. This method reduces the missed detection rate and false detection rate, improves the positioning accuracy, and meets the requirements of real-time detection of pedestrian objects.

Keywords: pedestrian detection, feature detection, convolutional neural network, real-time detection, YOLOv3

Procedia PDF Downloads 143

Authors: Raghavi C. Janaswamy

Abstract:

In the healthcare sector, heterogenous data elements such as patients, diagnosis, symptoms, conditions, observation text from physician notes, and prescriptions form the essentials of the Electronic Health Record (EHR). The data in the form of clear text and images are stored or processed in a relational format in most systems. However, the intrinsic structure restrictions and complex joins of relational databases limit the widespread utility. In this regard, the design and development of realistic mapping and deep connections as real-time objects offer unparallel advantages. Herein, a graph neural network-based classification of EHR data has been developed. The patient conditions have been predicted as a node classification task using a graph-based open source EHR data, Synthea Database, stored in Tigergraph. The Synthea DB dataset is leveraged due to its closer representation of the real-time data and being voluminous. The graph model is built from the EHR heterogeneous data using python modules, namely, pyTigerGraph to get nodes and edges from the Tigergraph database, PyTorch to tensorize the nodes and edges, PyTorch-Geometric (PyG) to train the Graph Neural Network (GNN) and adopt the self-supervised learning techniques with the AutoEncoders to generate the node embeddings and eventually perform the node classifications using the node embeddings. The model predicts patient conditions ranging from common to rare situations. The outcome is deemed to open up opportunities for data querying toward better predictions and accuracy.

Keywords: electronic health record, graph neural network, heterogeneous data, prediction

Procedia PDF Downloads 87

Authors: C. Senabre, S. Valero, M. Lopez, E. Velasco, M. Sanchez

Abstract:

This paper focuses on an application of linear mixed models to short-term load forecasting. The challenge of this research is to improve a currently working model at the Spanish Transport System Operator, programmed by us, and based on linear autoregressive techniques and neural networks. The forecasting system currently forecasts each of the regions within the Spanish grid separately, even though the behavior of the load in each region is affected by the same factors in a similar way. A load forecasting system has been verified in this work by using the real data from a utility. In this research it has been used an integration of several regions into a linear mixed model as starting point to obtain the information from other regions. Firstly, the systems to learn general behaviors present in all regions, and secondly, it is identified individual deviation in each regions. The technique can be especially useful when modeling the effect of special days with scarce information from the past. The three most relevant regions of the system have been used to test the model, focusing on special day and improving the performance of both currently working models used as benchmark. A range of comparisons with different forecasting models has been conducted. The forecasting results demonstrate the superiority of the proposed methodology.

Keywords: short-term load forecasting, mixed effects models, neural networks, mixed effects models

Procedia PDF Downloads 192

Authors: Chen Xi, Lao Xuerui, Li Junjie, Jiang Yike, Wang Hanwei, Zeng Zihao

Abstract:

To further promote the development of smart cities, the microscopic "nerve endings" of the City Intelligent Model (CIM) are extended to be more sensitive. In this paper, we develop a pedestrian trajectory twin terminal based on the CIM and CNN technology. It also uses 5G networks, architectural and geoinformatics technologies, convolutional neural networks, combined with deep learning networks for human behaviour recognition models, to provide empirical data such as 'pedestrian flow data and human behavioural characteristics data', and ultimately form spatial performance evaluation criteria and spatial performance warning systems, to make the empirical data accurate and intelligent for prediction and decision making.

Keywords: urban planning, urban governance, CIM, artificial intelligence, convolutional neural network

Procedia PDF Downloads 154

Authors: Magdi M. Nabi, Ding-Li Yu

Abstract:

This paper presents a neural network based model predictive control (MPC) strategy to control a strongly exothermic reaction with complicated nonlinear kinetics given by Chylla-Haase polymerization reactor that requires a very precise temperature control to maintain product uniformity. In the benchmark scenario, the operation of the reactor must be guaranteed under various disturbing influences, e.g., changing ambient temperatures or impurity of the monomer. Such a process usually controlled by conventional cascade control, it provides a robust operation, but often lacks accuracy concerning the required strict temperature tolerances. The predictive control strategy based on the RBF neural model is applied to solve this problem to achieve set-point tracking of the reactor temperature against disturbances. The result shows that the RBF based model predictive control gives reliable result in the presence of some disturbances and keeps the reactor temperature within a tight tolerance range around the desired reaction temperature.

Keywords: Chylla-Haase reactor, RBF neural network modelling, model predictive control, semi-batch reactors

Procedia PDF Downloads 468

Authors: R. B. Knudsen, O. T. Rasmussen, R. A. Alphinas

Abstract:

The reason for conducting this research is to develop an algorithm that is capable of classifying news articles from the automobile industry, according to the competitive actions that they entail, with the use of Text Mining (TM) methods. It is needed to test how to properly preprocess the data for this research by preparing pipelines which fits each algorithm the best. The pipelines are tested along with nine different classification algorithms in the realm of regression, support vector machines, and neural networks. Preliminary testing for identifying the optimal pipelines and algorithms resulted in the selection of two algorithms with two different pipelines. The two algorithms are Logistic Regression (LR) and Artificial Neural Network (ANN). These algorithms are optimized further, where several parameters of each algorithm are tested. The best result is achieved with the ANN. The final model yields an accuracy of 0.79, a precision of 0.80, a recall of 0.78, and an F1 score of 0.76. By removing three of the classes that created noise, the final algorithm is capable of reaching an accuracy of 94%.

Keywords: Artificial Neural network, Competitive dynamics, Logistic Regression, Text classification, Text mining

Procedia PDF Downloads 122

Authors: Shengbiao Wu, Weihua Cao, Min Wu, Can Liu

Abstract:

This paper presents an intelligent tuning method of microwave filter based on complex neural network and improved space mapping. The tuning process consists of two stages: the initial tuning and the fine tuning. At the beginning of the tuning, the return loss of the filter is transferred to the passband via the error of phase. During the fine tuning, the phase shift caused by the transmission line and the higher order mode is removed by the curve fitting. Then, an Cauchy method based on the admittance parameter (Y-parameter) is used to extract the coupling matrix. The influence of the resonant cavity loss is eliminated during the parameter extraction process. By using processed data pairs (the amount of screw variation and the variation of the coupling matrix), a tuning model is established by the complex neural network. In view of the improved space mapping algorithm, the mapping relationship between the actual model and the ideal model is established, and the amplitude and direction of the tuning is constantly updated. Finally, the tuning experiment of the eight order coaxial cavity filter shows that the proposed method has a good effect in tuning time and tuning precision.

Keywords: microwave filter, scattering parameter, coupling matrix, intelligent tuning

Procedia PDF Downloads 315

Authors: A. Bayle, F. Plouraboué

Abstract:

Water Distribution Network (WDN) still suffers from a lack of knowledge about fast pressure transient events prediction, although the latter may considerably impact their durability. Accidental or planned operating activities indeed give rise to complex pressure interactions and may drastically modified the local pressure value generating leaks and, in rare cases, pipe’s break. In this context, a numerical predictive analysis is conducted to prevent such event and optimize network management. A couple of Python/FORTRAN 90, home-made software, has been developed using Method Of Characteristic (MOC) solving for water-hammer equations. The solver is validated by direct comparison with theoretical and experimental measurement in simple configurations whilst afterward extended to network analysis. The algorithm's most costly steps are designed for parallel computation. A various set of boundary conditions and energetic losses models are considered for the network simulations. The results are analyzed in both real and frequencies domain and provide crucial information on the pressure distribution behavior within the network.

Keywords: energetic losses models, method of characteristic, numerical predictive analysis, water distribution network, water hammer

Procedia PDF Downloads 236

Authors: Jui-Chen Huang, Shou-Hsiung Cheng

Abstract:

This study examined their willingness to use telecare for people who have had experience falling in the last three months in Taiwan. This study adopted convenience sampling and a structural questionnaire to collect data. It was based on the definition and the constructs related to the Health Belief Model (HBM). HBM is comprised of seven constructs: perceived benefits (PBs), perceived disease threat (PDT), perceived barriers of taking action (PBTA), external cues to action (ECUE), internal cues to action (ICUE), attitude toward using (ATT), and behavioral intention to use (BI). This study adopted Fuzzy Neural Network (FNN) to put forward an effective method. It shows the dependence of ATT on PB, PDT, PBTA, ECUE, and ICUE. The training and testing data RMSE (root mean square error) are 0.028 and 0.166 in the FNN, respectively. The training and testing data RMSE are 0.828 and 0.578 in the regression model, respectively. On the other hand, as to the dependence of ATT on BI, as presented in the FNN, the training and testing data RMSE are 0.050 and 0.109, respectively. The training and testing data RMSE are 0.529 and 0.571 in the regression model, respectively. The results show that the FNN method is better than the regression analysis. It is an effective and viable good way.

Keywords: fall, fuzzy neural network, health belief model, telecare, willingness

Procedia PDF Downloads 202

Authors: K. Padmavathi, K. Sri Ramakrishna

Abstract:

This study describes a procedure for the detection of Left and Right Bundle Branch Block (LBBB and RBBB) ECG patterns using spectral Coherence(SC) technique and LM Neural Network. The Coherence function finds common frequencies between two signals and evaluate the similarity of the two signals. The QT variations of Bundle Blocks are observed in lead V1 of ECG. Spectral Coherence technique uses Welch method for calculating PSD. For the detection of normal and Bundle block beats, SC output values are given as the input features for the LMNN classifier. Overall accuracy of LMNN classifier is 99.5 percent. The data was collected from MIT-BIH Arrhythmia database.

Keywords: bundle block, SC, LMNN classifier, welch method, PSD, MIT-BIH, arrhythmia database

Procedia PDF Downloads 283

Authors: Halimatu S. Abdullahi, Ray E. Sheriff, Fatima Mahieddine

Abstract:

Agriculture is essential to the continuous existence of human life as they directly depend on it for the production of food. The exponential rise in population calls for a rapid increase in food with the application of technology to reduce the laborious work and maximize production. Technology can aid/improve agriculture in several ways through pre-planning and post-harvest by the use of computer vision technology through image processing to determine the soil nutrient composition, right amount, right time, right place application of farm input resources like fertilizers, herbicides, water, weed detection, early detection of pest and diseases etc. This is precision agriculture which is thought to be solution required to achieve our goals. There has been significant improvement in the area of image processing and data processing which has being a major challenge. A database of images is collected through remote sensing, analyzed and a model is developed to determine the right treatment plans for different crop types and different regions. Features of images from vegetations need to be extracted, classified, segmented and finally fed into the model. Different techniques have been applied to the processes from the use of neural network, support vector machine, fuzzy logic approach and recently, the most effective approach generating excellent results using the deep learning approach of convolution neural network for image classifications. Deep Convolution neural network is used to determine soil nutrients required in a plantation for maximum production. The experimental results on the developed model yielded results with an average accuracy of 99.58%.

Keywords: convolution, feature extraction, image analysis, validation, precision agriculture

Procedia PDF Downloads 318

Authors: René A. Sørensen, Søren Skovsen, Peter Christiansen, Henrik Karstoft

Abstract:

Autonomous agricultural machines act in stochastic surroundings and therefore, must be able to perceive the surroundings in real time. This perception can be achieved using image sensors combined with advanced machine learning, in particular Deep Learning. Deep convolutional neural networks excel in labeling and perceiving color images and since the cost of high-quality RGB-cameras is low, the hardware cost of good perception depends heavily on memory and computation power. This paper investigates the possibility of designing lightweight convolutional neural networks for semantic segmentation (pixel wise classification) with reduced hardware requirements, to allow for embedded usage in autonomous agricultural machines. Using compression techniques, a lightweight convolutional neural network is designed to perform real-time semantic segmentation on an embedded platform. The network is trained on two large datasets, ImageNet and Pascal Context, to recognize up to 400 individual classes. The 400 classes are remapped into agricultural superclasses (e.g. human, animal, sky, road, field, shelterbelt and obstacle) and the ability to provide accurate real-time perception of agricultural surroundings is studied. The network is applied to the case of autonomous grass mowing using the NVIDIA Tegra X1 embedded platform. Feeding case-specific images to the network results in a fully segmented map of the superclasses in the image. As the network is still being designed and optimized, only a qualitative analysis of the method is complete at the abstract submission deadline. Proceeding this deadline, the finalized design is quantitatively evaluated on 20 annotated grass mowing images. Lightweight convolutional neural networks for semantic segmentation can be implemented on an embedded platform and show competitive performance with regards to accuracy and speed. It is feasible to provide cost-efficient perceptive capabilities related to semantic segmentation for autonomous agricultural machines.

Keywords: autonomous agricultural machines, deep learning, safety, visual perception

Procedia PDF Downloads 398

Authors: Oluwafunmibi Omotayo Fasanya, Augustine Kena Adjei

Abstract:

Diabetic nephropathy (DN) is a major complication of diabetes, with fibrosis and propionate metabolism playing critical roles in its progression. Identifying biomarkers linked to these pathways may provide novel insights into DN diagnosis and treatment. This study aims to identify biomarkers associated with fibrosis and propionate metabolism in DN. Analyze the biological pathways and regulatory mechanisms of these biomarkers. Develop a machine learning model to predict DN-related biomarkers and validate their functional roles. Publicly available transcriptome datasets related to DN (GSE96804 and GSE104948) were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/gds), and 924 propionate metabolism-related genes (PMRGs) and 656 fibrosis-related genes (FRGs) were identified. The analysis began with the extraction of DN-differentially expressed genes (DN-DEGs) and propionate metabolism-related DEGs (PM-DEGs), followed by the intersection of these with fibrosis-related genes to identify key intersected genes. Instead of relying on traditional models, we employed a combination of deep neural networks (DNNs) and ensemble methods such as Gradient Boosting Machines (GBM) and XGBoost to enhance feature selection and biomarker discovery. Recursive feature elimination (RFE) was coupled with these advanced algorithms to refine the selection of the most critical biomarkers. Functional validation was conducted using convolutional neural networks (CNN) for gene set enrichment and immunoinfiltration analysis, revealing seven significant biomarkers—SLC37A4, ACOX2, GPD1, ACE2, SLC9A3, AGT, and PLG. These biomarkers are involved in critical biological processes such as fatty acid metabolism and glomerular development, providing a mechanistic link to DN progression. Furthermore, a TF–miRNA–mRNA regulatory network was constructed using natural language processing models to identify 8 transcription factors and 60 miRNAs that regulate these biomarkers, while a drug–gene interaction network revealed potential therapeutic targets such as UROKINASE–PLG and ATENOLOL–AGT. This integrative approach, leveraging deep learning and ensemble models, not only enhances the accuracy of biomarker discovery but also offers new perspectives on DN diagnosis and treatment, specifically targeting fibrosis and propionate metabolism pathways.

Keywords: diabetic nephropathy, deep neural networks, gradient boosting machines (GBM), XGBoost

Procedia PDF Downloads 12

Authors: Hassan Masoumi, Ahad Salimi, Nazanin Barhemmat, Babak Gholami

Abstract:

Artificial neural networks have gained a lot of interest as empirical models for their powerful representational capacity, multi input and output mapping characteristics. In fact, most feed-forward networks with nonlinear nodal functions have been proved to be universal approximates. In this paper, we propose a new supervised method for color image classification based on self organizing feature maps (SOFM). This algorithm is based on competitive learning. The method partitions the input space using self-organizing feature maps to introduce the concept of local neighborhoods. Our image classification system entered into RGB image. Experiments with simulated data showed that separability of classes increased when increasing training time. In additional, the result shows proposed algorithms are effective for color image classification.

Keywords: classification, SOFM algorithm, neural network, neighborhood, RGB image

Procedia PDF Downloads 482

Authors: Zeina Bou Diab, Arij Daou

Abstract:

The neural mechanisms of sequential behaviors are intensively studied, with songbirds a focus for learned vocal production. We are studying the premotor nucleus HVC at a nexus of multiple pathways contributing to song learning and production. The HVC consists of multiple classes of neuronal populations, each has its own cellular, electrophysiological and functional properties. During singing, a large subset of motor cortex analog-projecting HVCRA neurons emit a single 6-10 ms burst of spikes at the same time during each rendition of song, a large subset of basal ganglia-projecting HVCX neurons fire 1 to 4 bursts that are similarly time locked to vocalizations, while HVCINT neurons fire tonically at average high frequency throughout song with prominent modulations whose timing in relation to song remains unresolved. This opens the opportunity to define models relating explicit HVC circuitry to how these neurons work cooperatively to control learning and singing. We developed conductance-based Hodgkin-Huxley models for the three classes of HVC neurons (based on the ion channels previously identified from in vitro recordings) and connected them in several physiologically realistic networks (based on the known synaptic connectivity and specific glutaminergic and gabaergic pharmacology) via different architecture patterning scenarios with the aim to replicate the in vivo firing patterning behaviors. We are able, through these networks, to reproduce the in vivo behavior of each class of HVC neurons, as shown by the experimental recordings. The different network architectures developed highlight different mechanisms that might be contributing to the propagation of sequential neural activity (continuous or punctate) in the HVC and to the distinctive firing patterns that each class exhibits during singing. Examples of such possible mechanisms include: 1) post-inhibitory rebound in HVCX and their population patterns during singing, 2) different subclasses of HVCINT interacting via inhibitory-inhibitory loops, 3) mono-synaptic HVCX to HVCRA excitatory connectivity, and 4) structured many-to-one inhibitory synapses from interneurons to projection neurons, and others. Replication is only a preliminary step that must be followed by model prediction and testing.

Keywords: computational modeling, neural networks, temporal neural sequences, ionic currents, songbird

Procedia PDF Downloads 72

Authors: Jaiden Xuan Schraut, Leon Liu, Yiqiao Yin

Abstract:

Computer vision in medical diagnosis has achieved a high level of success in diagnosing diseases with high accuracy. However, conventional classifiers that produce an image to-label result provides insufficient information for medical professionals to judge and raise concerns over the trust and reliability of a model with results that cannot be explained. In order to gain local insight into cancerous regions, separate tasks such as imaging segmentation need to be implemented to aid the doctors in treating patients, which doubles the training time and costs which renders the diagnosis system inefficient and difficult to be accepted by the public. To tackle this issue and drive AI-first medical solutions further, this paper proposes a multi-output network that follows a U-Net architecture for image segmentation output and features an additional convolutional neural networks (CNN) module for auxiliary classification output. Class activation maps are a method of providing insight into a convolutional neural network’s feature maps that leads to its classification but in the case of lung diseases, the region of interest is enhanced by U-net-assisted Class Activation Map (CAM) visualization. Therefore, our proposed model combines image segmentation models and classifiers to crop out only the lung region of a chest X-ray’s class activation map to provide a visualization that improves the explainability and is able to generate classification results simultaneously which builds trust for AI-led diagnosis systems. The proposed U-Net model achieves 97.61% accuracy and a dice coefficient of 0.97 on testing data from the COVID-QU-Ex Dataset which includes both diseased and healthy lungs.

Keywords: multi-output network model, U-net, class activation map, image classification, medical imaging analysis

Procedia PDF Downloads 205

Authors: Shuen-Tai Wang, Fang-An Kuo, Chau-Yi Chou, Yu-Bin Fang

Abstract:

2016 has become the year of the Artificial Intelligence explosion. AI technologies are getting more and more matured that most world well-known tech giants are making large investment to increase the capabilities in AI. Machine learning is the science of getting computers to act without being explicitly programmed, and deep learning is a subset of machine learning that uses deep neural network to train a machine to learn  features directly from data. Deep learning realizes many machine learning applications which expand the field of AI. At the present time, deep learning frameworks have been widely deployed on servers for deep learning applications in both academia and industry. In training deep neural networks, there are many standard processes or algorithms, but the performance of different frameworks might be different. In this paper we evaluate the running performance of two state-of-the-art distributed deep learning frameworks that are running training calculation in parallel over multi GPU and multi nodes in our cloud environment. We evaluate the training performance of the frameworks with ResNet-50 convolutional neural network, and we analyze what factors that result in the performance among both distributed frameworks as well. Through the experimental analysis, we identify the overheads which could be further optimized. The main contribution is that the evaluation results provide further optimization directions in both performance tuning and algorithmic design.

Keywords: artificial intelligence, machine learning, deep learning, convolutional neural networks

Procedia PDF Downloads 212

Authors: Paola Causin, Andrea Aspri, Alessandro Benfenati

Abstract:

Diffuse Optical Tomography (DOT) is an emergent medical imaging technique which employs NIR light to estimate the spatial distribution of optical coefficients in biological tissues for diagnostic purposes, in a noninvasive and non-ionizing manner. DOT reconstruction is a severely ill-conditioned problem due to prevalent scattering of light in the tissue. In this contribution, we present our research in adopting hybrid knowledgedriven/data-driven approaches which exploit the existence of well assessed physical models and build upon them neural networks integrating the availability of data. Namely, since in this context regularization procedures are mandatory to obtain a reasonable reconstruction [1], we explore the use of neural networks as tools to include prior information on the solution. 2. Materials and Methods The idea underlying our approach is to leverage neural networks to solve PDE-constrained inverse problems of the form 𝒒 ∗ = 𝒂𝒓𝒈 𝒎𝒊𝒏𝒒 𝐃(𝒚, 𝒚̃), (1) where D is a loss function which typically contains a discrepancy measure (or data fidelity) term plus other possible ad-hoc designed terms enforcing specific constraints. In the context of inverse problems like (1), one seeks the optimal set of physical parameters q, given the set of observations y. Moreover, 𝑦̃ is the computable approximation of y, which may be as well obtained from a neural network but also in a classic way via the resolution of a PDE with given input coefficients (forward problem, Fig.1 box ). Due to the severe ill conditioning of the reconstruction problem, we adopt a two-fold approach: i) we restrict the solutions (optical coefficients) to lie in a lower-dimensional subspace generated by auto-decoder type networks. This procedure forms priors of the solution (Fig.1 box ); ii) we use regularization procedures of type 𝒒̂ ∗ = 𝒂𝒓𝒈𝒎𝒊𝒏𝒒 𝐃(𝒚, 𝒚̃)+ 𝑹(𝒒), where 𝑹(𝒒) is a regularization functional depending on regularization parameters which can be fixed a-priori or learned via a neural network in a data-driven modality. To further improve the generalizability of the proposed framework, we also infuse physics knowledge via soft penalty constraints (Fig.1 box ) in the overall optimization procedure (Fig.1 box ). 3. Discussion and Conclusion DOT reconstruction is severely hindered by ill-conditioning. The combined use of data-driven and knowledgedriven elements is beneficial and allows to obtain improved results, especially with a restricted dataset and in presence of variable sources of noise.

Keywords: inverse problem in tomography, deep learning, diffuse optical tomography, regularization

Procedia PDF Downloads 75

Authors: Mustafa Al Sukar, Azzam Sleit, Abdullatif Abu-Dalhoum, Bassam Al-Kasasbeh

Abstract:

Use and abuse of drugs by teens is very common and can have dangerous consequences. The drugs contribute to physical and sexual aggression such as assault or rape. Some teenagers regularly use drugs to compensate for depression, anxiety or a lack of positive social skills. Teen resort to smoking should not be minimized because it can be "gateway drugs" for other drugs (marijuana, cocaine, hallucinogens, inhalants, and heroin). The combination of teenagers' curiosity, risk taking behavior, and social pressure make it very difficult to say no. This leads most teenagers to the questions: "Will it hurt to try once?" Nowadays, technological advances are changing our lives very rapidly and adding a lot of technologies that help us to track the risk of drug abuse such as smart phones, Wireless Sensor Networks (WSNs), Internet of Things (IoT), etc. This technique may help us to early discovery of drug abuse in order to prevent an aggravation of the influence of drugs on the abuser. In this paper, we have developed a Decision Support System (DSS) for detecting the drug abuse using Artificial Neural Network (ANN); we used a Multilayer Perceptron (MLP) feed-forward neural network in developing the system. The input layer includes 50 variables while the output layer contains one neuron which indicates whether the person is a drug addict. An iterative process is used to determine the number of hidden layers and the number of neurons in each one. We used multiple experiment models that have been completed with Log-Sigmoid transfer function. Particularly, 10-fold cross validation schemes are used to access the generalization of the proposed system. The experiment results have obtained 98.42% classification accuracy for correct diagnosis in our system. The data had been taken from 184 cases in Jordan according to a set of questions compiled from Specialists, and data have been obtained through the families of drug abusers.

Keywords: drug addiction, artificial neural networks, multilayer perceptron (MLP), decision support system

Procedia PDF Downloads 301

Authors: Lin Cheng, Zijiang Yang

Abstract:

Program synthesis is the task to automatically generate programs based on user specification. In this paper, we present a framework that synthesizes programs from flow charts that serve as accurate and intuitive specification. In order doing so, we propose a deep neural network called GRCNN that recognizes graph structure from its image. GRCNN is trained end-to-end, which can predict edge and node information of the flow chart simultaneously. Experiments show that the accuracy rate to synthesize a program is 66.4%, and the accuracy rates to recognize edge and node are 94.1% and 67.9%, respectively. On average, it takes about 60 milliseconds to synthesize a program.

Keywords: program synthesis, flow chart, specification, graph recognition, CNN

Procedia PDF Downloads 120

Authors: Hamidreza Ghazvinian, Khosro Ghazvinian, Touba Khodaiean

Abstract:

The importance of evaporation estimation in water resources and agricultural studies is undeniable. Pan evaporation are used as an indicator to determine the evaporation of lakes and reservoirs around the world due to the ease of interpreting its data. In this research, intelligent models were investigated in estimating pan evaporation on a daily basis. Shahroud and Mayamey were considered as the studied cities. These two cities are located in Semnan province in Iran. The mentioned cities have dry weather conditions that are susceptible to high evaporation potential. Meteorological data of 11 years of synoptic stations of Shahrood and Mayamey cities were used. The intelligent models used in this study are Artificial Neural Network (ANN), Least Squares Support Vector Machine (LSSVM), and M5 tree models. Meteorological parameters of minimum and maximum air temperature (Tmax, Tmin), wind speed (WS), sunshine hours (SH), air pressure (PA), relative humidity (RH) as selected input data and evaporation data from pan (EP) to The output data was considered. 70% of data is used at the education level, and 30 % of the data is used at the test level. Models used with explanation coefficient evaluation (R2) Root of Mean Squares Error (RMSE) and Mean Absolute Error (MAE). The results for the two Shahroud and Mayamey stations showed that the above three models' operations are rather appropriate.

Keywords: pan evaporation, intelligent methods, shahroud, mayamey

Procedia PDF Downloads 75

Authors: Neziha Jaouedi, Noureddine Boujnah, Mohamed Salim Bouhlel

Abstract:

In the framework of human machine interaction systems enhancement, we focus throw this paper on human behavior analysis and action recognition. Human behavior is characterized by actions and reactions duality (movements, psychological modification, verbal and emotional expression). It’s worth noting that many information is hidden behind gesture, sudden motion points trajectories and speeds, many research works reconstructed an information retrieval issues. In our work we will focus on motion extraction, tracking and action recognition using wavelet network approaches. Our contribution uses an analysis of human subtraction by Gaussian Mixture Model (GMM) and body movement through trajectory models of motion constructed from kalman filter. These models allow to remove the noise using the extraction of the main motion features and constitute a stable base to identify the evolutions of human activity. Each modality is used to recognize a human action using wavelets of derived beta distributions approach. The proposed approach has been validated successfully on a subset of KTH and UCF sports database.

Keywords: feautures extraction, human action classifier, wavelet neural network, beta wavelet

Procedia PDF Downloads 411

Authors: Lee Jeong Min, Lee Mi Hee, Eo Yang Dam

Abstract:

Geospatial feature extraction is a very important issue in the remote sensing research. In the meantime, the image classification based on statistical techniques, but, in recent years, data mining and machine learning techniques for automated image processing technology is being applied to remote sensing it has focused on improved results generated possibility. In this study, artificial neural network and decision tree technique is applied to classify the high-resolution satellite images, as compared to the MLC processing result is a statistical technique and an analysis of the pros and cons between each of the techniques.

Keywords: remote sensing, artificial neural network, decision tree, maximum likelihood classification

Procedia PDF Downloads 347

Authors: Dalin Si, Azizan Aziz, Bertrand Lasternas

Abstract:

To encourage building owners to purchase electricity at the wholesale market and reduce building peak demand, this study aims to develop models that predict day-ahead hourly electricity consumption and demand using artificial neural network (ANN) and support vector machine (SVM). All prediction models are built in Python, with tool Scikit-learn and Pybrain. The input data for both consumption and demand prediction are time stamp, outdoor dry bulb temperature, relative humidity, air handling unit (AHU), supply air temperature and solar radiation. Solar radiation, which is unavailable a day-ahead, is predicted at first, and then this estimation is used as an input to predict consumption and demand. Models to predict consumption and demand are trained in both SVM and ANN, and depend on cooling or heating, weekdays or weekends. The results show that ANN is the better option for both consumption and demand prediction. It can achieve 15.50% to 20.03% coefficient of variance of root mean square error (CVRMSE) for consumption prediction and 22.89% to 32.42% CVRMSE for demand prediction, respectively. To conclude, the presented models have potential to help building owners to purchase electricity at the wholesale market, but they are not robust when used in demand response control.

Keywords: building energy prediction, data mining, demand response, electricity market

Procedia PDF Downloads 317