Search results for: graph convolutional networks (GCNs)

Authors: Osung Im, Neha Yadav, Eui Hoon Lee, Joong Hoon Kim

Abstract:

Artificial Neural Network (ANN) approach is commonly used in lots of fields for forecasting. In water resources engineering, forecast of water level or inflow of reservoir is useful for various kind of purposes. Due to advantages of ANN, many papers were written for inflow prediction in river networks, but in this study, ANN is used in urban sewer networks. The growth of severe rain storm in Korea has increased flood damage severely, and the precipitation distribution is getting more erratic. Therefore, effective pump operation in pump station is an essential task for the reduction in urban area. If real time inflow of pump station reservoir can be predicted, it is possible to operate pump effectively for reducing the flood damage. This study used ANN model for pump station reservoir inflow prediction using upstream sewer depth data. For this study, rainfall events, sewer depth, and inflow into Banpo pump station reservoir between years of 2013-2014 were considered. Feed – Forward Back Propagation (FFBF), Cascade – Forward Back Propagation (CFBP), Elman Back Propagation (EBP) and Nonlinear Autoregressive Exogenous (NARX) were used as ANN model for prediction. A comparison of results with ANN model suggests that ANN is a powerful tool for inflow prediction using the sewer depth data.

Keywords: artificial neural network, forecasting, reservoir inflow, sewer depth

Procedia PDF Downloads 317

Authors: Mst Shapna Akter, Hossain Shahriar

Abstract:

Neural network approaches are machine learning methods used in many domains, such as healthcare and cyber security. Neural networks are mostly known for dealing with image datasets. While training with the images, several fundamental mathematical operations are carried out in the Neural Network. The operation includes a number of algebraic and mathematical functions, including derivative, convolution, and matrix inversion and transposition. Such operations require higher processing power than is typically needed for computer usage. Central Processing Unit (CPU) is not appropriate for a large image size of the dataset as it is built with serial processing. While Graphics Processing Unit (GPU) has parallel processing capabilities and, therefore, has higher speed. This paper uses advanced Neural Network techniques such as VGG16, Resnet50, Densenet, Inceptionv3, Xception, Mobilenet, XGBOOST-VGG16, and our proposed models to compare CPU and GPU resources. A system for classifying autism disease using face images of an autistic and non-autistic child was used to compare performance during testing. We used evaluation matrices such as Accuracy, F1 score, Precision, Recall, and Execution time. It has been observed that GPU runs faster than the CPU in all tests performed. Moreover, the performance of the Neural Network models in terms of accuracy increases on GPU compared to CPU.

Keywords: autism disease, neural network, CPU, GPU, transfer learning

Procedia PDF Downloads 118

Authors: Yuhong Zhang, Dan Lu, Chenyang Bu, Peipei Li, Kui Yu, Xindong Wu

Abstract:

The Cross-Lingual Entity Alignment (CLEA) task aims to find the aligned entities that refer to the same identity from two knowledge graphs (KGs) in different languages. It is an effective way to enhance the performance of data mining for KGs with scarce resources. In real-world applications, the neighborhood structures of the same entities in different KGs tend to be non-isomorphic, which makes the representation of entities contain diverse semantic information and then poses a great challenge for CLEA. In this paper, we try to address this challenge from two perspectives. On the one hand, the cross-KG relation completion rules are designed with the alignment constraint of entities and relations to improve the topology isomorphism of two KGs. On the other hand, a representation method combining isomorphic weights is designed to include more isomorphic semantics for counterpart entities, which will benefit the CLEA. Experiments show that our model can improve the isomorphism of two KGs and the alignment performance, especially for two non-isomorphic KGs.

Keywords: knowledge graphs, cross-lingual entity alignment, non-isomorphic, relation completion

Procedia PDF Downloads 123

Authors: Muhammad Zubair Khan, Yugyung Lee

Abstract:

Deep learning has recently achieved enormous response in semantic image segmentation. The previously developed U-Net inspired architectures operate with continuous stride and pooling operations, leading to spatial data loss. Also, the methods lack establishing long-term pixels connection to preserve context knowledge and reduce spatial loss in prediction. This article developed encoder-decoder architecture with bi-directional LSTM embedded in long skip-connections and densely connected convolution blocks. The network non-linearly combines the feature maps across encoder-decoder paths for finding dependency and correlation between image pixels. Additionally, the densely connected convolutional blocks are kept in the final encoding layer to reuse features and prevent redundant data sharing. The method applied batch-normalization for reducing internal covariate shift in data distributions. The empirical evidence shows a promising response to our method compared with other semantic segmentation techniques.

Keywords: deep learning, semantic segmentation, image analysis, pixels connection, convolution neural network

Procedia PDF Downloads 102

Authors: Elena Carcano, James Ball

Abstract:

This work details the strategies adopted by the Italian Water Utilities during the distribution of water in emergency conditions which glide from earthquakes and droughts to floods and fires. Several water bureaus located over the national territory have been interviewed, and the collected information has been used in a database of potential interventions to be taken. The work discusses the actions adopted by water utilities. These are generally prioritized in order to minimize the social, temporal, and economic burden that the damaged and nearby areas need to support. Actions are defined relying on the Bayesian Network Approach, which constitutes the hard core of any decision support system. The Bayesian Networks give answers to interventions to real and most likely risky cases. The added value of this research consists in supplying the National Bureau, namely Protezione Civile, in charge of managing havoc and catastrophic situations with a univocal plot outline so as to be able to handle actions uniformly at the expense of different local laws or contradictory customs which squander any recovery conditions, proper technical service, and economic aids. The paper is organized as follows: in section 1, the introduction is stated; section 2 provides a brief discussion of BNNs (Bayesian Networks), section 3 introduces the adopted methodology; and in the last sections, results are presented, and conclusions are drawn.

Keywords: hierarchical process, strategic plan, water emergency conditions, water supply

Procedia PDF Downloads 160

Authors: Pranav Gunhal., Krish Jhurani

Abstract:

This study tackles the significant computational challenge of predicting excited state dynamics in organic photovoltaic (OPV) materials—a pivotal factor in the performance of solar energy solutions. Time-dependent density functional theory (TDDFT), though effective, is computationally prohibitive for larger and more complex molecules. As a solution, the research explores the application of transformer neural networks, a type of artificial intelligence (AI) model known for its superior performance in natural language processing, to predict excited state dynamics in OPV materials. The methodology involves a two-fold process. First, the transformer model is trained on an extensive dataset comprising over 10,000 TDDFT calculations of excited state dynamics from a diverse set of OPV materials. Each training example includes a molecular structure and the corresponding TDDFT-calculated excited state lifetimes and key electronic transitions. Second, the trained model is tested on a separate set of molecules, and its predictions are rigorously compared to independent TDDFT calculations. The results indicate a remarkable degree of predictive accuracy. Specifically, for a test set of 1,000 OPV materials, the transformer model predicted excited state lifetimes with a mean absolute error of 0.15 picoseconds, a negligible deviation from TDDFT-calculated values. The model also correctly identified key electronic transitions contributing to the excited state dynamics in 92% of the test cases, signifying a substantial concordance with the results obtained via conventional quantum chemistry calculations. The practical integration of the transformer model with existing quantum chemistry software was also realized, demonstrating its potential as a powerful tool in the arsenal of materials scientists and chemists. The implementation of this AI model is estimated to reduce the computational cost of predicting excited state dynamics by two orders of magnitude compared to conventional TDDFT calculations. The successful utilization of transformer neural networks to accurately predict excited state dynamics provides an efficient computational pathway for the accelerated discovery and design of new OPV materials, potentially catalyzing advancements in the realm of sustainable energy solutions.

Keywords: transformer neural networks, organic photovoltaic materials, excited state dynamics, time-dependent density functional theory, predictive modeling

Procedia PDF Downloads 118

Authors: Alper Taner, Yesim Benal Oztekin, Huseyin Duran

Abstract:

In this study, an Artificial Neural Network (ANN) was developed in order to classify barley varieties. For this purpose, physical properties of barley varieties were determined and ANN techniques were used. The physical properties of 8 barley varieties grown in Turkey, namely thousand kernel weight, geometric mean diameter, sphericity, kernel volume, surface area, bulk density, true density, porosity and colour parameters of grain, were determined and it was found that these properties were statistically significant with respect to varieties. As ANN model, three models, N-l, N-2 and N-3 were constructed. The performances of these models were compared. It was determined that the best-fit model was N-1. In the N-1 model, the structure of the model was designed to be 11 input layers, 2 hidden layers and 1 output layer. Thousand kernel weight, geometric mean diameter, sphericity, kernel volume, surface area, bulk density, true density, porosity and colour parameters of grain were used as input parameter; and varieties as output parameter. R2, Root Mean Square Error and Mean Error for the N-l model were found as 99.99%, 0.00074 and 0.009%, respectively. All results obtained by the N-l model were observed to have been quite consistent with real data. By this model, it would be possible to construct automation systems for classification and cleaning in flourmills.

Keywords: physical properties, artificial neural networks, barley, classification

Procedia PDF Downloads 178

Authors: Malik Muhammad Arslan, Muneeb Ullah, Dai Shihan, Faizan Khan, Xiaodong Yang

Abstract:

Internet Service Providers are driving endless demands for higher bandwidth and data throughput as new services and applications require higher bandwidth. Users want immediate and accurate data delivery. This article focuses on converting old conventional networks into passive optical networks based on time division and wavelength division multiplexing. The main focus of this research is to use a hybrid of time-division multiplexing and wavelength-division multiplexing to improve network efficiency and performance. In this paper, we design an 80 Gbps Passive Optical Network (PON), which meets the need of the Next Generation PON Stage 2 (NGPON2) proposed in this paper. The hybrid of the Time and Wavelength division multiplexing (TWDM) is said to be the best solution for the implementation of NGPON2, according to Full-Service Access Network (FSAN). To co-exist with or replace the current PON technologies, many wavelengths of the TWDM can be implemented simultaneously. By utilizing 8 pairs of wavelengths that are multiplexed and then transmitted over optical fiber for 40 Kms and on the receiving side, they are distributed among 256 users, which shows that the solution is reliable for implementation with an acceptable data rate. From the results, it can be concluded that the overall performance, Quality Factor, and bandwidth of the network are increased, and the Bit Error rate is minimized by the integration of this approach.

Keywords: bit error rate, fiber to the home, passive optical network, time and wavelength division multiplexing

Procedia PDF Downloads 70

Authors: Panaya Sudta, Wanchalerm Patanacharoenwong, Prachya Bumrungkun

Abstract:

As With the evolution of smart grids, ensuring the reliability and efficiency of electrical systems in smart cities has become crucial. This paper proposes a distinct approach that combines advanced machine learning techniques to accurately predict electrical failures and address voltage reliability issues. This approach aims to improve the accuracy and efficiency of reliability evaluations in smart cities. The aim of this research is to develop a comprehensive predictive model that accurately predicts electrical failures and voltage reliability in smart cities. This model integrates RFM analysis, K-means clustering, and LSTM networks to achieve this objective. The research utilizes RFM analysis, traditionally used in customer value assessment, to categorize and analyze electrical components based on their failure recency, frequency, and monetary impact. K-means clustering is employed to segment electrical components into distinct groups with similar characteristics and failure patterns. LSTM networks are used to capture the temporal dependencies and patterns in customer data. This integration of RFM, K-means, and LSTM results in a robust predictive tool for electrical failures and voltage reliability. The proposed model has been tested and validated on diverse electrical utility datasets. The results show a significant improvement in prediction accuracy and reliability compared to traditional methods, achieving an accuracy of 92.78% and an F1-score of 0.83. This research contributes to the proactive maintenance and optimization of electrical infrastructures in smart cities. It also enhances overall energy management and sustainability. The integration of advanced machine learning techniques in the predictive model demonstrates the potential for transforming the landscape of electrical system management within smart cities. The research utilizes diverse electrical utility datasets to develop and validate the predictive model. RFM analysis, K-means clustering, and LSTM networks are applied to these datasets to analyze and predict electrical failures and voltage reliability. The research addresses the question of how accurately electrical failures and voltage reliability can be predicted in smart cities. It also investigates the effectiveness of integrating RFM analysis, K-means clustering, and LSTM networks in achieving this goal. The proposed approach presents a distinct, efficient, and effective solution for predicting and mitigating electrical failures and voltage issues in smart cities. It significantly improves prediction accuracy and reliability compared to traditional methods. This advancement contributes to the proactive maintenance and optimization of electrical infrastructures, overall energy management, and sustainability in smart cities.

Keywords: electrical state prediction, smart grids, data-driven method, long short-term memory, RFM, k-means, machine learning

Procedia PDF Downloads 56

Authors: Ali Berkan Ural

Abstract:

This paper presents the analysis, evaluation, and pre-diagnosis of early stage breast based diagnostic problems (breast cancer, nodulesorlumps) by Computer Aided Diagnosing (CAD) system from mammogram radiological images. According to the statistics, the time factor is crucial to discover the disease in the patient (especially in women) as possible as early and fast. In the study, a new algorithm is developed using advanced image processing and deep learning method to detect and classify the problem at earlystagewithmoreaccuracy. This system first works with image processing methods (Image acquisition, Noiseremoval, Region Growing Segmentation, Morphological Operations, Breast BorderExtraction, Advanced Segmentation, ObtainingRegion Of Interests (ROIs), etc.) and segments the area of interest of the breast and then analyzes these partly obtained area for cancer detection/lumps in order to diagnosis the disease. After segmentation, with using the Spectrogramimages, 5 different deep learning based methods (specified Convolutional Neural Network (CNN) basedAlexNet, ResNet50, VGG16, DenseNet, Xception) are applied to classify the breast based problems.

Keywords: computer aided diagnosis, breast cancer, region growing, segmentation, deep learning

Procedia PDF Downloads 95

Authors: Zhou Luchen, Wu Yubing, Burra Venkata Durga Kumar

Abstract:

As the scale of the network becomes larger and more complex than before, the density of user devices is also increasing. The development of Unmanned Aerial Vehicle (UAV) networks is able to collect and transform data in an efficient way by using software-defined networks (SDN) technology. This paper proposed a three-layer distributed and dynamic cluster architecture to manage UAVs by using an AI-based resource allocation calculation algorithm to address the overloading network problem. Through separating services of each UAV, the UAV hierarchical cluster system performs the main function of reducing the network load and transferring user requests, with three sub-tasks including data collection, communication channel organization, and data relaying. In this cluster, a head node and a vice head node UAV are selected considering the Central Processing Unit (CPU), operational (RAM), and permanent (ROM) memory of devices, battery charge, and capacity. The vice head node acts as a backup that stores all the data in the head node. The k-means clustering algorithm is used in order to detect high load regions and form the UAV layered clusters. The whole process of detecting high load areas, forming and selecting UAV clusters, and moving the selected UAV cluster to that area is proposed as offloading traffic algorithm.

Keywords: k-means, resource allocation, SDN, UAV network, unmanned aerial vehicles

Procedia PDF Downloads 111

Authors: Ganesh Nandakumaran, Mehmet Karaata

Abstract:

A wave is a distributed execution, often made up of a broadcast phase followed by a feedback phase, requiring the participation of all the system processes before a particular event called decision is taken. Wave algorithms with one initiator such as the 1-wave algorithm have been shown to be very efficient for broadcasting messages in tree networks. Extensions of this algorithm broadcasting a sequence of waves using a single initiator have been implemented in algorithms such as the m-wave algorithm. However as the network size increases, having a single initiator adversely affects the message delivery times to nodes further away from the initiator. As a remedy, broadcast waves can be allowed to be initiated by multiple initiator nodes distributed across the network to reduce the completion time of broadcasts. These waves initiated by one or more initiator processes form a collection of waves covering the entire network. Solutions to global-snapshots, distributed broadcast and various synchronization problems can be solved efficiently using waves with multiple concurrent initiators. In this paper, we propose the first stabilizing multi-wave sequence algorithm implementing waves started by multiple initiator processes such that every process in the network receives at least one sequence of broadcasts. Due to being stabilizing, the proposed algorithm can withstand transient faults and do not require initialization. We view a fault as a transient fault if it perturbs the configuration of the system but not its program.

Keywords: distributed computing, multi-node broadcast, propagation of information with feedback and cleaning (PFC), stabilization, wave algorithms

Procedia PDF Downloads 504

Authors: Siming Xie

Abstract:

In social networks, the term homophily refers to the tendency of agents with similar characteristics to link with one another and is so robustly observed across many contexts and dimensions. The starting point of my research is the observation that the “type” of agents is not a single exogenous variable. Agents, despite their differences in race, religion, and other hard to alter characteristics, may share interests and engage in activities that cut across those predetermined lines. This research aims to capture the interactions of homophily effects in a model where agents have two-dimension characteristics (i.e., race and personal hobbies such as basketball, which one either likes or dislikes) and with biases in meeting opportunities and in favor of same-type friendships. A novel feature of my model is providing a matching process with biased meeting probability on different dimensions, which could help to understand the structuring process in multidimensional networks without missing layer interdependencies. The main contribution of this study is providing a welfare based matching process for agents with multi-dimensional characteristics. In particular, this research shows that the biases in meeting opportunities on one dimension would lead to the emergence of homophily on the other dimension. The objective of this research is to determine the pattern of homophily in network formations, which will shed light on our understanding of segregation and its remedies. By constructing a two-dimension matching process, this study explores a method to describe agents’ homophilous behavior in a social network with multidimension and construct a game in which the minorities and majorities play different strategies in a society. It also shows that the optimal strategy is determined by the relative group size, where society would suffer more from social segregation if the two racial groups have a similar size. The research also has political implications—cultivating the same characteristics among agents helps diminishing social segregation, but only if the minority group is small enough. This research includes both theoretical models and empirical analysis. Providing the friendship formation model, the author first uses MATLAB to perform iteration calculations, then derives corresponding mathematical proof on previous results, and last shows that the model is consistent with empirical evidence from high school friendships. The anonymous data comes from The National Longitudinal Study of Adolescent Health (Add Health).

Keywords: homophily, multidimension, social networks, friendships

Procedia PDF Downloads 170

Authors: Maria Isabel Garcia-Planas, Maria Victoria Garcia-Camba

Abstract:

Recent advances in cognitive neuroscience have allowed a step forward in perceiving the processes involved in learning from the point of view of the acquisition of new information or the modification of existing mental content. The evoked potentials technique reveals how basic brain processes interact to achieve adequate and flexible behaviours. The objective of this work, using evoked potentials, is to study if it is possible to distinguish if a patient suffers a specific type of learning disorder to decide the possible therapies to follow. The methodology used, is the analysis of the dynamics of different areas of the brain during a cognitive activity to find the relationships between the different areas analyzed in order to better understand the functioning of neural networks. Also, the latest advances in neuroscience have revealed the existence of different brain activity in the learning process that can be highlighted through the use of non-invasive, innocuous, low-cost and easy-access techniques such as, among others, the evoked potentials that can help to detect early possible neuro-developmental difficulties for their subsequent assessment and cure. From the study of the amplitudes and latencies of the evoked potentials, it is possible to detect brain alterations in the learning process specifically in dyscalculia, to achieve specific corrective measures for the application of personalized psycho pedagogical plans that allow obtaining an optimal integral development of the affected people.

Keywords: dyscalculia, neurodevelopment, evoked potentials, Learning disabilities, neural networks

Procedia PDF Downloads 140

Authors: Giselle Maggie-Fer Castañeda Lozano

Abstract:

The main objective of this research was to explore the role of neural network architectures in simulating behavioral phenomena as a potential explanation for selective associations, specifically related to biological constraints on learning. Biological constraints on learning refer to the limitations observed in conditioning procedures, where learning is expected to occur. The study involved simulations of five different experiments exploring various phenomena and sources of biological constraints in learning. These simulations included the interaction between response and reinforcer, stimulus and reinforcer, specificity of stimulus-reinforcer associations, species differences, neuroanatomical constraints, and learning in uncontrolled conditions. The overall results demonstrated that by manipulating neural network architectures, conditions can be created to model and explain diverse biological constraints frequently reported in comparative psychology literature as learning typicities. Additionally, the simulations offer predictive content worthy of experimental testing in the pursuit of new discoveries regarding the specificity of learning. The implications and limitations of these findings are discussed. Finally, it is suggested that this research could inaugurate a line of inquiry involving the use of neural networks to study biological factors in behavior, fostering the development of more ethical and precise research practices.

Keywords: comparative psychology, connectionism, conditioning, experimental analysis of behavior, neural networks

Procedia PDF Downloads 71

Authors: Niousha Bagheri Khulenjani, Mohammad Saniee Abadeh

Abstract:

Learning from very big datasets is a significant problem for most present data mining and machine learning algorithms. MicroRNA (miRNA) is one of the important big genomic and non-coding datasets presenting the genome sequences. In this paper, a hybrid method for the classification of the miRNA data is proposed. Due to the variety of cancers and high number of genes, analyzing the miRNA dataset has been a challenging problem for researchers. The number of features corresponding to the number of samples is high and the data suffer from being imbalanced. The feature selection method has been used to select features having more ability to distinguish classes and eliminating obscures features. Afterward, a Convolutional Neural Network (CNN) classifier for classification of cancer types is utilized, which employs a Genetic Algorithm to highlight optimized hyper-parameters of CNN. In order to make the process of classification by CNN faster, Graphics Processing Unit (GPU) is recommended for calculating the mathematic equation in a parallel way. The proposed method is tested on a real-world dataset with 8,129 patients, 29 different types of tumors, and 1,046 miRNA biomarkers, taken from The Cancer Genome Atlas (TCGA) database.

Keywords: cancer classification, feature selection, deep learning, genetic algorithm

Procedia PDF Downloads 111

Authors: Cheng Fang, Lingwei Quan, Cunyue Lu

Abstract:

Human pose estimation and tracking are to accurately identify and locate the positions of human joints in the video. It is a computer vision task which is of great significance for human motion recognition, behavior understanding and scene analysis. There has been remarkable progress on human pose estimation in recent years. However, more researches are needed for human pose tracking especially for online tracking. In this paper, a framework, called PoseSRPN, is proposed for online single-person pose estimation and tracking. We use Siamese network attaching a pose estimation branch to incorporate Single-person Pose Tracking (SPT) and Visual Object Tracking (VOT) into one framework. The pose estimation branch has a simple network structure that replaces the complex upsampling and convolution network structure with deconvolution. By augmenting the loss of fully convolutional Siamese network with the pose estimation task, pose estimation and tracking can be trained in one stage. Once trained, PoseSRPN only relies on a single bounding box initialization and producing human joints location. The experimental results show that while maintaining the good accuracy of pose estimation on COCO and PoseTrack datasets, the proposed method achieves a speed of 59 frame/s, which is superior to other pose tracking frameworks.

Keywords: computer vision, pose estimation, pose tracking, Siamese network

Procedia PDF Downloads 153

Authors: K. N. Seetharamu, Naveen Teggi, Kiranakumar Dhavalagi, Narayana Kamath

Abstract:

A method of temperature calculations is developed to study the conditions leading to hot spot occurrence on multicore chips. A physical model which has salient features of multicore chips is incorporated for the analysis. The model consists of active and background cell laid out in a checkered pattern, and this pattern repeats itself in each fine grain active cells. The die has three layers i) body ii) buried oxide layer iii) wiring layer, stacked one above the other with heat source placed at the interface between wiring and buried oxide layer. With this model we propose analytical method to calculate the target hotspot temperature, heat flow to top and bottom layers of the die and thermal resistance components at each granularity level, assuming appropriate values of die dimensions and parameters. Finally we attempt to find an easier method for the calculation of the target hotspot temperature using graph.

Keywords: checkered pattern, granularity level, heat conduction, multicore chips, target hotspot temperature

Procedia PDF Downloads 466

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

Abstract:

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

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

Procedia PDF Downloads 207

Authors: Rajvir Kaur, Jeewani Anupama Ginige

Abstract:

With recent trends in Big Data and advancements in Information and Communication Technologies, the healthcare industry is at the stage of its transition from clinician oriented to technology oriented. Many people around the world die of cancer because the diagnosis of disease was not done at an early stage. Nowadays, the computational methods in the form of Machine Learning (ML) are used to develop automated decision support systems that can diagnose cancer with high confidence in a timely manner. This paper aims to carry out the comparative evaluation of a selected set of ML classifiers on two existing datasets: breast cancer and cervical cancer. The ML classifiers compared in this study are Decision Tree (DT), Support Vector Machine (SVM), k-Nearest Neighbor (k-NN), Logistic Regression, Ensemble (Bagged Tree) and Artificial Neural Networks (ANN). The evaluation is carried out based on standard evaluation metrics Precision (P), Recall (R), F1-score and Accuracy. The experimental results based on the evaluation metrics show that ANN showed the highest-level accuracy (99.4%) when tested with breast cancer dataset. On the other hand, when these ML classifiers are tested with the cervical cancer dataset, Ensemble (Bagged Tree) technique gave better accuracy (93.1%) in comparison to other classifiers.

Keywords: artificial neural networks, breast cancer, classifiers, cervical cancer, f-score, machine learning, precision, recall

Procedia PDF Downloads 277

Authors: Matej Komanec, Jan Bohata, Stanislav Zvanovec, Tomas Nemecek, Jan Broucek, Josef Beran

Abstract:

With the exponential growth of social networks, video streaming and increasing demands on data rates, the number of newly built data centers rises proportionately. The data centers, however, have to adjust to the rapidly increased amount of data that has to be processed. For this purpose, multi-mode (MM) fiber based infrastructures are often employed. It stems from the fact, the connections in data centers are typically realized within a short distance, and the application of MM fibers and components considerably reduces costs. On the other hand, the usage of MM components brings specific requirements for installation service conditions. Moreover, it has to be taken into account that MM fiber components have a higher production tolerance for parameters like core and cladding diameters, eccentricity, etc. Due to the high demands for the reliability of data center components, the determination of properly excited optical field inside the MM fiber core belongs to the key parameters while designing such an MM optical system architecture. Appropriately excited mode field of the MM fiber provides optimal power budget in connections, leads to the decrease of insertion losses (IL) and achieves effective modal bandwidth (EMB). The main parameter, in this case, is the encircled flux (EF), which should be properly defined for variable optical sources and consequent different mode-field distribution. In this paper, we present detailed investigation and measurements of the mode field distribution for short MM links purposed in particular for data centers with the emphasis on reliability and safety. These measurements are essential for large MM network design. The various scenarios, containing different fibers and connectors, were tested in terms of IL and mode-field distribution to reveal potential challenges. Furthermore, we focused on estimation of particular defects and errors, which can realistically occur like eccentricity, connector shifting or dust, were simulated and measured, and their dependence to EF statistics and functionality of data center infrastructure was evaluated. The experimental tests were performed at two wavelengths, commonly used in MM networks, of 850 nm and 1310 nm to verify EF statistics. Finally, we provide recommendations for data center systems and networks, using OM3 and OM4 MM fiber connections.

Keywords: optical fiber, multi-mode, data centers, encircled flux

Procedia PDF Downloads 375

Authors: Bandar Alahmadi, Lethia Jackson

Abstract:

Machine Learning model is used today in many real-life applications. The safety and security of such model is important, so the results of the model are as accurate as possible. One challenge of machine learning model security is the adversarial examples attack. Adversarial examples are designed by the attacker to cause the machine learning model to misclassify the input. We propose a method to generate adversarial examples to attack image classifiers. We are modifying the successfully classified images, so a classifier misclassifies them after the modification. In our method, we do not update the whole image, but instead we detect the important region, modify it, place it back to the original image, and then run it through a classifier. The algorithm modifies the detected region using two methods. First, it will add abstract image matrix on back of the detected image matrix. Then, it will perform a rotation attack to rotate the detected region around its axes, and embed the trace of image in image background. Finally, the attacked region is placed in its original position, from where it was removed, and a smoothing filter is applied to smooth the background with foreground. We test our method in cascade classifier, and the algorithm is efficient, the classifier confident has dropped to almost zero. We also try it in CNN (Convolutional neural network) with higher setting and the algorithm was successfully worked.

Keywords: adversarial examples, attack, computer vision, image processing

Procedia PDF Downloads 339

Authors: A. A. Balkhi, G. M. Mir, Javid A. Sheikh

Abstract:

To cater the demands of increasing traffic with new applications the cellular mobile networks face new changes in deployment in infrastructure for making cellular networks heterogeneous. To reduce overhead processing the densely deployed cells require smart behavior with self-organizing capabilities with high adaptation to the neighborhood. We propose self-organization of unused resources usually excessive unused channels of neighbouring cells with densely populated cells to reduce handover failure rates. The neighboring cells share unused channels after fulfilling some conditional candidature criterion using threshold values so that they are not suffered themselves for starvation of channels in case of any abrupt change in traffic pattern. The cells are classified as ‘red’, ‘yellow’, or ‘green’, as per the available channels in cell which is governed by traffic pattern and thresholds. To combat the deficiency of channels in red cell, migration of unused channels from under-loaded cells, hierarchically from the qualified candidate neighboring cells is explored. The resources are returned back when the congested cell is capable of self-contained traffic management. In either of the cases conditional sharing of resources is executed for enhanced traffic management so that User Equipment (UE) is provided uninterrupted services with high Quality of Service (QoS). The fuzzy logic-based simulation results show that the proposed algorithm is efficiently in coincidence with improved successful handoffs.

Keywords: candidate cell, channel sharing, fuzzy logic, handover, small cells

Procedia PDF Downloads 120

Authors: Peter Sauer, Thomas Hinze, Petra Hofstedt

Abstract:

To avoid obstacles in the surrounding environment and to navigate to a given target belong to the most important tasks for mobile robots. According to these tasks different data structures are suitable. To avoid near obstacles, occupancy grid maps are an ideal representation of the surroundings. For less fine grained tasks, such as navigating from one room to another in an apartment, pure grid maps are inappropriate. Grid maps are very detailed, calculating paths to navigate between rooms based on grid maps would take too long. Instead, graph-based data structures, so-called topologies, turn out to be a proper choice for such tasks. In this paper we present two methods to dynamically create topologies from grid maps. Both methods are based on non-uniform grid maps. The topologies are generated on-the-fly and can easily be modified to represent changes in the environment. This allows a hybrid approach to control mobile robots, where, depending on the situation and the current task, either the grid map or the generated topology may be used.

Keywords: robot navigation, occupancy grids, topological maps, dynamic map creation

Procedia PDF Downloads 563

Authors: Vidhu Mitha Goutham, Ruma Mukherjee

Abstract:

The low availability of well-trained, unlimited, dynamic-access models for specific languages makes it hard for corporate users to adopt quick translation techniques and incorporate them into product solutions. As translation tasks increasingly require a dynamic sequence learning curve; stable, cost-free opensource models are scarce. We survey and compare current translation techniques and propose a modified sequence to sequence model repurposed with attention techniques. Sequence learning using an encoder-decoder model is now paving the path for higher precision levels in translation. Using a Convolutional Neural Network (CNN) encoder and a Recurrent Neural Network (RNN) decoder background, we use Fairseq tools to produce an end-to-end bilingually trained Spanish-English machine translation model including source language detection. We acquire competitive results using a duo-lingo-corpus trained model to provide for prospective, ready-made plug-in use for compound sentences and document translations. Our model serves a decent system for large, organizational data translation needs. While acknowledging its shortcomings and future scope, it also identifies itself as a well-optimized deep neural network model and solution.

Keywords: attention, encoder-decoder, Fairseq, Seq2Seq, Spanish, translation

Procedia PDF Downloads 175

Authors: Messaoud Eljamai, Sami Hidouri

Abstract:

Massive multiple-input multiple-output (MIMO) is an emerging technology for new cellular networks such as 5G systems. Its principle is to use many antennas per cell in order to maximize the network's spectral efficiency. Inter-cellular interference remains a fundamental problem. The use of massive MIMO will not derogate from the rule. It improves performances only when the number of antennas is significantly greater than the number of users. This, considerably, limits the networks spectral efficiency. In this paper, a coordinated detector for an uplink massive MIMO system is proposed in order to mitigate the inter-cellular interference. The proposed scheme combines the coordinated multipoint technique with an interference-cancelling algorithm. It requires the serving cell to send their received symbols, after processing, decision and error detection, to the interfered cells via a backhaul link. Each interfered cell is capable of eliminating intercellular interferences by generating and subtracting the user’s contribution from the received signal. The resulting signal is more reliable than the original received signal. This allows the uplink massive MIMO system to improve their performances dramatically. Simulation results show that the proposed detector improves system spectral efficiency compared to classical linear detectors.

Keywords: massive MIMO, COMP, interference canceling algorithm, spectral efficiency

Procedia PDF Downloads 147

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: Xiang Yu, Chuntao Feng, Lu Yang, Meiyang Song, Wenhao Zhou

Abstract:

The existing two-dimensional micro-Doppler features extraction ignores the correlation information between the spatial and temporal dimension features. For the range-Doppler map, the time dimension is introduced, and a frequency modulation continuous wave (FMCW) radar human fall detection algorithm based on time-varying range-Doppler features is proposed. Firstly, the range-Doppler sequence maps are generated from the echo signals of the continuous motion of the human body collected by the radar. Then the three-dimensional data cube composed of multiple frames of range-Doppler maps is input into the three-dimensional Convolutional Neural Network (3D CNN). The spatial and temporal features of time-varying range-Doppler are extracted by the convolution layer and pool layer at the same time. Finally, the extracted spatial and temporal features are input into the fully connected layer for classification. The experimental results show that the proposed fall detection algorithm has a detection accuracy of 95.66%.

Keywords: FMCW radar, fall detection, 3D CNN, time-varying range-doppler features

Procedia PDF Downloads 122

Authors: Zoran Šverko, Saša Vlahinić, Nino Stojković, Ivan Markovinović

Abstract:

In this article, the classification of target letters is performed using data from the EEG P300 Speller paradigm. Neural networks trained with the results of dynamic connectivity analysis between different brain regions are used for classification. Dynamic connectivity analysis is based on the adaptive window size and the imaginary part of the complex Pearson correlation coefficient. Brain dynamics are analysed using the relative intersection of confidence intervals for the imaginary component of the complex Pearson correlation coefficient method (RICI-imCPCC). The RICI-imCPCC method overcomes the shortcomings of currently used dynamical connectivity analysis methods, such as the low reliability and low temporal precision for short connectivity intervals encountered in constant sliding window analysis with wide window size and the high susceptibility to noise encountered in constant sliding window analysis with narrow window size. This method overcomes these shortcomings by dynamically adjusting the window size using the RICI rule. This method extracts information about brain connections for each time sample. Seventy percent of the extracted brain connectivity information is used for training and thirty percent for validation. Classification of the target word is also done and based on the same analysis method. As far as we know, through this research, we have shown for the first time that dynamic connectivity can be used as a parameter for classifying EEG signals.

Keywords: dynamic connectivity analysis, EEG, neural networks, Pearson correlation coefficients

Procedia PDF Downloads 214

Authors: Bahareh Mostofian

Abstract:

Foreign Direct Investment (FDI) has long been recognized as a crucial driver of economic growth and development in less-developed countries and their integration into Global Production Networks (GPNs). FDI not only brings capital from the core countries but also technology, innovation, and know-how knowledge that can upgrade the capabilities of host automotive industries. On the other hand, FDI can also have negative impacts on host countries if it leads to significant import dependency. In the case of the Iranian automotive sector, the industry greatly benefited from FDI, with Western carmakers dominating the market. Over time, various types of know-how knowledge, including joint ventures (JVs), trade licenses, and technical assistance, have been provided, helping Iran upgrade its automotive industry. While after the severe geopolitical obstacles imposed by both the EU and the U.S., the industry became over-reliant on the car and spare parts imports, and the lack of emphasis on knowledge transfer further affected the growth and development of the Iranian automotive sector. To address these challenges, current research has adopted a descriptive-analytical methodology to illustrate the gradual changes accrued with foreign suppliers through FDI. The research finding shows that after the two-phase imposed sanctions, the detrimental linkages created by overreliance on the car and spare parts imports without any industrial upgrading negatively affected the growth and development of the national and assembled products of the Iranian automotive sector.

Keywords: less-developed country, FDI, GPNs, automotive industry, Iran

Procedia PDF Downloads 73