Search results for: recurrent neural network

Authors: Manoj Kumar Dutta

Abstract:

Wavelength division multiplexing (WDM) technology is the most promising technology for the proper utilization of huge raw bandwidth provided by an optical fiber. One of the key problems in implementing the all-optical WDM network is the packet contention. This problem can be solved by several different techniques. In time domain approach the packet contention can be reduced by incorporating fiber delay lines (FDLs) as optical buffer in the switch architecture. Different types of buffering architectures are reported in literatures. In the present paper a comparative performance analysis of three most popular FDL architectures are presented in order to obtain the best contention resolution performance. The analysis is further extended to consider the effect of different fiber non-linearities on the network performance.

Keywords: WDM network, contention resolution, optical buffering, non-linearity, throughput

Procedia PDF Downloads 451

Authors: Jarek Krajewski, Andre Wittenborn, Martin Sauerland

Abstract:

This abstract describes a promising approach for estimating post-traumatic stress disorder (PTSD) based on prosodic speech characteristics. It illustrates the validity of this method by briefly discussing results from an Arabic refugee sample (N= 47, 32 m, 15 f). A well-established standardized self-report scale “Reaction of Adolescents to Traumatic Stress” (RATS) was used to determine the ground truth level of PTSD. The speech material was prompted by telling about autobiographical related sadness inducing experiences (sampling rate 16 kHz, 8 bit resolution). In order to investigate PTSD-induced speech changes, a self-developed set of 136 prosodic speech features was extracted from the .wav files. This set was adapted to capture traumatization related speech phenomena. An artificial neural network (ANN) machine learning model was applied to determine the PTSD level and reached a correlation of r = .37. These results indicate that our classifiers can achieve similar results to those seen in speech-based stress research.

Keywords: speech prosody, PTSD, machine learning, feature extraction

Procedia PDF Downloads 90

Authors: Kamel Allaw, Jocelyne Adjizian Gerard, Makram Chehayeb, Nada Badaro Saliba

Abstract:

In developing countries, such as Lebanon, the demographic data are hardly available due to the absence of the mechanization of population system. The aim of this study is to evaluate, using only remote sensing data, the correlations between the number of population and the characteristics of roads network (length of primary roads, length of secondary roads, total length of roads, density and percentage of roads and the number of intersections). In order to find the influence of the different factors on the demographic data, we studied the degree of correlation between each factor and the number of population. The results of this study have shown a strong correlation between the number of population and the density of roads and the number of intersections.

Keywords: population, road network, statistical correlations, remote sensing

Procedia PDF Downloads 161

Authors: Gunasekaran Raja, Ramkumar Jayaraman

Abstract:

In this paper, we consider CCL-N (Cooperative Cross Layer Network) topology based on the cross layer (both centralized and distributed) environment to form network communities. Various performance metrics related to the IEEE 802.16 networks are discussed to design CCL-N Topology. In CCL-N topology, nodes are classified as master nodes (Master Base Station [MBS]) and serving nodes (Relay Station [RS]). Nodes communities are organized based on the networking terminologies. Based on CCL-N Topology, various simulation analyses for both transparent and non-transparent relays are tabulated and throughput efficiency is calculated. Weighted load balancing problem plays a challenging role in IEEE 802.16 network. CoTS (Concurrent Transmission Scheduling) Scheme is formulated in terms of three aspects – transmission mechanism based on identical communities, different communities and identical node communities. CoTS scheme helps in identifying the weighted load balancing problem. Based on the analytical results, modularity value is inversely proportional to that of the error value. The modularity value plays a key role in solving the CoTS problem based on hop count. The transmission mechanism for identical node community has no impact since modularity value is same for all the network groups. In this paper three aspects of communities based on the modularity value which helps in solving the problem of weighted load balancing and CoTS are discussed.

Keywords: cross layer network topology, concurrent scheduling, modularity value, network communities and weighted load balancing

Procedia PDF Downloads 265

Authors: Lim Teck Beng, Thiha Kyaw, Poh Boon Kiat, Lee Ngai Meng

Abstract:

This work presents discussion on the possibility of merging two emerging technologies in microwave; wireless power transfer (WPT) and RF energy harvesting. The current state of art of the two technologies is discussed and the strength and weakness of the two technologies is also presented. The equivalent circuit of wireless power transfer is modeled and explained as how the range and efficiency can be further increased by controlling certain parameters in the receiver. The different techniques of harvesting the RF energy from the ambient are also extensive study. Last but not least, we demonstrate that a low power wireless sensor network (WSN) can be power up by RF energy harvesting. The WSN is designed to transmit every 3 minutes of information containing the temperature of the environment and also the voltage of the node. One thing worth mention is both the sensors that are used for measurement are also powering up by the RF energy harvesting system.

Keywords: energy harvesting, wireless power transfer, wireless sensor network and magnetic coupled resonator

Procedia PDF Downloads 519

Authors: Mohammad Karimi Moridani, Mohammad Abdi Zadeh, Zahra Shahiazar Mazraeh

Abstract:

In recent years, the use of intelligent systems in biomedical engineering has increased dramatically, especially in the diagnosis of various diseases. Also, due to the relatively simple recording of the electrocardiogram signal (ECG), this signal is a good tool to show the function of the heart and diseases associated with it. The aim of this paper is to design an intelligent system for automatically detecting a normal electrocardiogram signal from abnormal one. Using this diagnostic system, it is possible to identify a person's heart condition in a very short time and with high accuracy. The data used in this article are from the Physionet database, available in 2016 for use by researchers to provide the best method for detecting normal signals from abnormalities. Data is of both genders and the data recording time varies between several seconds to several minutes. All data is also labeled normal or abnormal. Due to the low positional accuracy and ECG signal time limit and the similarity of the signal in some diseases with the normal signal, the heart rate variability (HRV) signal was used. Measuring and analyzing the heart rate variability with time to evaluate the activity of the heart and differentiating different types of heart failure from one another is of interest to the experts. In the preprocessing stage, after noise cancelation by the adaptive Kalman filter and extracting the R wave by the Pan and Tampkinz algorithm, R-R intervals were extracted and the HRV signal was generated. In the process of processing this paper, a new idea was presented that, in addition to using the statistical characteristics of the signal to create a return map and extraction of nonlinear characteristics of the HRV signal due to the nonlinear nature of the signal. Finally, the artificial neural networks widely used in the field of ECG signal processing as well as distinctive features were used to classify the normal signals from abnormal ones. To evaluate the efficiency of proposed classifiers in this paper, the area under curve ROC was used. The results of the simulation in the MATLAB environment showed that the AUC of the MLP and SVM neural network was 0.893 and 0.947, respectively. As well as, the results of the proposed algorithm in this paper indicated that the more use of nonlinear characteristics in normal signal classification of the patient showed better performance. Today, research is aimed at quantitatively analyzing the linear and non-linear or descriptive and random nature of the heart rate variability signal, because it has been shown that the amount of these properties can be used to indicate the health status of the individual's heart. The study of nonlinear behavior and dynamics of the heart's neural control system in the short and long-term provides new information on how the cardiovascular system functions, and has led to the development of research in this field. Given that the ECG signal contains important information and is one of the common tools used by physicians to diagnose heart disease, but due to the limited accuracy of time and the fact that some information about this signal is hidden from the viewpoint of physicians, the design of the intelligent system proposed in this paper can help physicians with greater speed and accuracy in the diagnosis of normal and patient individuals and can be used as a complementary system in the treatment centers.

Keywords: neart rate variability, signal processing, linear and non-linear features, classification methods, ROC Curve

Procedia PDF Downloads 262

Authors: Viktor Müller, Ulman Lindenberger

Abstract:

Neurophysiological evidence suggests that the physiological states of the system are characterized by specific network structures and network topology dynamics, demonstrating a robust interplay between network topology and function. It is also evident that interpersonal action coordination or social interaction (e.g., playing music in duets or groups) requires strong intra- and interbrain synchronization resulting in a specific hyper brain network activity across two or more brains to support such coordination or interaction. Such complex hyper brain networks can be described as multiplex or multilayer networks that have a specific multidimensional or multilayer network organization characteristic for superordinate systems and their constituents. The aim of the study was to describe multilayer hyper brain networks and synchronization patterns of guitarists playing guitar in a quartet by using electroencephalography (EEG) hyper scanning (simultaneous EEG recording from multiple brains) and following time-frequency decomposition and multilayer network construction, where within-frequency coupling (WFC) represents communication within different layers, and cross-frequency coupling (CFC) depicts communication between these layers. Results indicate that communication or coupling dynamics, both within and between the layers across the brains of the guitarists, play an essential role in action coordination and are particularly enhanced during periods of high demands on musical coordination. Moreover, multilayer hyper brain network topology and dynamical structure of guitar sounds showed specific guitar-guitar, brain-brain, and guitar-brain causal associations, indicating multilevel dynamics with upward and downward causation, contributing to the superordinate system dynamics and hyper brain functioning. It is concluded that the neuronal dynamics during interpersonal interaction are brain-wide and frequency-specific with the fine-tuned balance between WFC and CFC and can best be described in terms of multilayer multi-brain networks with specific network topology and connectivity strengths. Further sophisticated research is needed to deepen our understanding of these highly interesting and complex phenomena.

Keywords: EEG hyper scanning, intra- and interbrain coupling, multilayer hyper brain networks, social interaction, within- and cross-frequency coupling

Procedia PDF Downloads 72

Authors: P. R. Dushantha Chaminda, Peng Kai

Abstract:

In this review paper, we enclose the thought of wireless ad hoc networks and particularly mobile ad hoc network (MANET), their field of study, intention, concern, benefit and disadvantages, modifications, with relation of AODV routing protocol. Mobile computing is developing speedily with progression in wireless communications and wireless networking protocols. Making communication easy, we function most wireless network devices and sensor networks, movable, battery-powered, thus control on a highly constrained energy budget. However, progress in battery technology presents that only little improvements in battery volume can be expected in the near future. Moreover, recharging or substitution batteries is costly or unworkable, it is preferable to support energy waste level of devices low.

Keywords: wireless ad hoc network, energy efficient routing protocols, AODV, EOAODV, AODVEA, AODVM, AOMDV, FF-AOMDV, AOMR-LM

Procedia PDF Downloads 214

Authors: Yossiri Ariyakul, Piyakiat Insom, Poonyawat Sangiamkulthavorn, Takamichi Nakamoto

Abstract:

In this paper, the method of optimization algorithm for sensor network comprised of movable sensor nodes which can be used for odor source localization was proposed. A sensor node is composed of an odor sensor, an anemometer, and a wireless communication module. The odor intensity measured from the sensor nodes are sent to the processor to perform the localization based on optimization algorithm by which the odor source localization map is obtained as a result. The map can represent the exact position of the odor source or show the direction toward it remotely. The proposed method was experimentally validated by creating the odor source localization map using three, four, and five sensor nodes in which the accuracy to predict the position of the odor source can be observed.

Keywords: odor sensor, odor source localization, optimization, sensor network

Procedia PDF Downloads 299

Authors: Reza Alikhani

Abstract:

This research focuses on the establishment of horizontal cooperation among companies to enhance their operational efficiency and competitiveness. The study proposes an approach to horizontal collaboration, called coalition configuration, which involves partnering companies sharing distribution centers in a network design problem. The paper investigates which coalition should be formed in each distribution center to minimize the total cost of the network. Moreover, potential uncertainties, such as operational and disruption risks, are considered during the collaborative design phase. To address this problem, a two-stage robust optimization model for collaborative distribution network design under surging demand and facility disruptions is presented, along with a column-and-constraint generation algorithm to obtain exact solutions tailored to the proposed formulation. Extensive numerical experiments are conducted to analyze solutions obtained by the model in various scenarios, including decisions ranging from fully centralized to fully decentralized settings, collaborative versus non-collaborative approaches, and different amounts of uncertainty budgets. The results show that the coalition formation mechanism proposes some solutions that are competitive with the savings of the grand coalition. The research also highlights that collaboration increases network flexibility and resilience while reducing costs associated with demand and capacity uncertainties.

Keywords: logistics, warehouse sharing, robust facility location, collaboration for resilience

Procedia PDF Downloads 69

Authors: Venugopal Kummamuru

Abstract:

Corporate Governance (CG) is of utmost importance for running a company ethically. It is essential for the growth and success of the corporation. It is intended to increase the accountability of an organization to the larger context of the business environment. The general principles of CG include and are related to Shareholder recognition, Stakeholder interests, and focus on Corporate Social Responsibility (CSR), Clear Board responsibilities, Ethical behavior, and Business transparency. Network Marketing Organizations (NMOs) focus on marketing through direct-sales using people who are associated with the organization but are not their employees. This paper tries to study the importance of Ethics and CSR in an NMO and suggest a basic guideline for CG in NMO(s). This paper could be used as a basis or starting point for conducting an in-depth research to understand the difference in CG practices between NMO(s) and other organizations and define a standard set of guidelines for CG practice.

Keywords: corporate governance, corporate responsibility, direct selling, network marketing

Procedia PDF Downloads 318

Authors: Sun Lim, Kyun Jung

Abstract:

In this paper, we describe the control strategy of high speed parallel robot system with EtherCAT network. This work deals the parallel robot system with centralized control on the real-time operating system such as window TwinCAT3. Most control scheme and algorithm is implemented master platform on the PC, the input and output interface is ported on the slave side. The data is transferred by maximum 20usecond with 1000byte. EtherCAT is very high speed and stable industrial network. The control strategy with EtherCAT is very useful and robust on Ethernet network environment. The developed parallel robot is controlled pre-design nonlinear controller for 6G/0.43 cycle time of pick and place motion tracking. The experiment shows the good design and validation of the controller.

Keywords: parallel robot control, etherCAT, nonlinear control, parallel robot inverse kinematic

Procedia PDF Downloads 571

Authors: M. Hilani, B. Nassih

Abstract:

Phishing is one of the oldest and best known scams on the Internet. It can be defined as any type of telecommunications fraud that uses social engineering tricks to obtain confidential data from its victims. It’s a cybercrime aimed at stealing your sensitive information. Phishing is generally done via private email, so scammers impersonate large companies or other trusted entities to encourage victims to voluntarily provide information such as login credentials or, worse yet, credit card numbers. The COVID-19 theme is used by cybercriminals in multiple malicious campaigns like phishing. In this environment, messaging filtering solutions have become essential to protect devices that will now be used outside of the secure perimeter. Despite constantly updating methods to avoid these cyberattacks, the end result is currently insufficient. Many researchers are looking for optimal solutions to filter phishing emails, but we still need good results. In this work, we concentrated on solving the problem of detecting phishing emails using the different steps of NLP preprocessing, and we proposed and trained a model using one-dimensional CNN. Our study results show that our model obtained an accuracy of 99.99%, which demonstrates how well our model is working.

Keywords: phishing, e-mail, NLP preprocessing, CNN, e-mail filtering

Procedia PDF Downloads 126

Authors: Xiyang Li, Qi Yu, Lun Zhang

Abstract:

In the face of natural disasters and other emergency situations, determining the optimal location of rescue centers is crucial for improving rescue efficiency and minimizing impact on affected populations. This paper proposes a method that integrates genetic algorithms (GA) and backpropagation neural networks (BPNN) to address the site selection optimization problem for emergency rescue centers. We utilize BPNN to accurately estimate the cost of delivering supplies from rescue centers to each temporary camp. Moreover, a genetic algorithm with a special partially matched crossover (PMX) strategy is employed to ensure that the number of temporary camps assigned to each rescue center adheres to predetermined limits. Using the population distribution data during the 2022 epidemic in Jiading District, Shanghai, as an experimental case, this paper verifies the effectiveness of the proposed method. The experimental results demonstrate that the BPNN-GA method proposed in this study outperforms existing algorithms in terms of computational efficiency and optimization performance. Especially considering the requirements for computational resources and response time in emergency situations, the proposed method shows its ability to achieve rapid convergence and optimal performance in the early and mid-stages. Future research could explore incorporating more real-world conditions and variables into the model to further improve its accuracy and applicability.

Keywords: emergency rescue centers, genetic algorithms, back-propagation neural networks, site selection optimization

Procedia PDF Downloads 85

Authors: Navab Karimi, Tohid Alizadeh

Abstract:

An intelligent vision-based system was designed to measure the quality and purity of raisins. A machine vision setup was utilized to capture the images of bulk raisins in ranges of 5-50% mixed pure-impure berries. The textural features of bulk raisins were extracted using Grey-level Histograms, Co-occurrence Matrix, and Local Binary Pattern (a total of 108 features). Genetic Algorithm and neural network regression were used for selecting and ranking the best features (21 features). As a result, the GLCM features set was found to have the highest accuracy (92.4%) among the other sets. Followingly, multiple feature combinations of the previous stage were fed into the second regression (linear regression) to increase accuracy, wherein a combination of 16 features was found to be the optimum. Finally, a Support Vector Machine (SVM) classifier was used to differentiate the mixtures, producing the best efficiency and accuracy of 96.2% and 97.35%, respectively.

Keywords: sun-dried organic raisin, genetic algorithm, feature extraction, ann regression, linear regression, support vector machine, south azerbaijan.

Procedia PDF Downloads 73

Authors: Roshani S. Golhar, Neerajkumar S. Sathawane, Snehal Dongre

Abstract:

Background and objectives: As the, cardiovascular illnesses increasing and becoming cause of mortality worldwide, killing around lot of people each year. Arrhythmia is a type of cardiac illness characterized by a change in the linearity of the heartbeat. The goal of this study is to develop novel deep learning algorithms for successfully interpreting arrhythmia using a single second segment. Because the ECG signal indicates unique electrical heart activity across time, considerable changes between time intervals are detected. Such variances, as well as the limited number of learning data available for each arrhythmia, make standard learning methods difficult, and so impede its exaggeration. Conclusions: The proposed method was able to outperform several state-of-the-art methods. Also proposed technique is an effective and convenient approach to deep learning for heartbeat interpretation, that could be probably used in real-time healthcare monitoring systems

Keywords: electrocardiogram, ECG classification, neural networks, convolutional neural networks, portable document format

Procedia PDF Downloads 69

Authors: A. Fayad, Q. Alqhazaly, T. Cinkler

Abstract:

In light of the substantial increase in end-user requirements and the incessant need of network operators to upgrade the capabilities of access networks, in this paper, the performance of the different modulation formats on eight-channels Time and Wavelength Division Multiplexing Passive Optical Network (TWDM-PON) transmission system has been examined and compared. Limitations and features of modulation formats have been determined to outline the most suitable design to enhance the data rate and transmission reach to obtain the best performance of the network. The considered modulation formats are On-Off Keying Non-Return-to-Zero (NRZ-OOK), Carrier Suppressed Return to Zero (CSRZ), Duo Binary (DB), Modified Duo Binary (MODB), Quadrature Phase Shift Keying (QPSK), and Differential Quadrature Phase Shift Keying (DQPSK). The performance has been analyzed by varying transmission distances and bit rates under different channel spacing. Furthermore, the system is evaluated in terms of minimum Bit Error Rate (BER) and Quality factor (Qf) without applying any dispersion compensation technique, or any optical amplifier. Optisystem software was used for simulation purposes.

Keywords: BER, DuoBinary, NRZ-OOK, TWDM-PON

Procedia PDF Downloads 149

Authors: Juhi Faridi, Mohd. Ajmal Kafeel

Abstract:

The advancements in the field of Artificial Intelligence (AI) and technology has led to an evolution of an intelligent era. Neural networks, having the computational power and learning ability similar to the brain is one of the key AI technologies. Neuromorphic computing system (NCS) consists of the synaptic device, neuronal circuit, and neuromorphic architecture. Memristor are a promising candidate for neuromorphic computing systems, but when it comes to neuromorphic computing, the conductance behavior of the synaptic memristor or neuronal memristor needs to be studied thoroughly in order to fathom the neuroscience or computer science. Furthermore, there is a need of more simulation work for utilizing the existing device properties and providing guidance to the development of future devices for different performance requirements. Hence, development of NCS needs more simulation work to make use of existing device properties. This work aims to provide an insight to build neuronal circuits using memristors to achieve a Memristor based NCS.  Here we throw a light on the research conducted in the field of memristors for building analog and digital circuits in order to motivate the research in the field of NCS by building memristor based neural circuits for advanced AI applications. This literature is a step in the direction where we describe the various Key findings about memristors and its analog and digital circuits implemented over the years which can be further utilized in implementing the neuronal circuits in the NCS. This work aims to help the electronic circuit designers to understand how the research progressed in memristors and how these findings can be used in implementing the neuronal circuits meant for the recent progress in the NCS.

Keywords: analog circuits, digital circuits, memristors, neuromorphic computing systems

Procedia PDF Downloads 174

Authors: Noura Shehab-Eldeen, Mohamed Elsherbeeny, Hossam Elmetwally, Mohamed Salama, Ahmed Lotfy, Mohamed Elgamal, Hussein Sheashaa, Mohamed Sobh

Abstract:

Mitochondrial toxins including papaverine may be implicated in the etiology and pathogenesis of Parkinson's disease. The aim was to detect the effect of papaverine on the proliferation and viability of neural stem cells. Rat neural progenitor cells were isolated from embryos (E14) brains. The dispersed tissues were allowed to settle, then, The supernatant was centrifuged at 1,000 g for 5 min. The pellet was placed in Hank’s solution cultured as free-floating neurospheres Dulbecco’s modified Eagle medium (DMEM) and Hams F12 (3:1) supplemented with B27 (Invitrogen GmBH, Karlsruhe, Germany), 20 ng/mL epidermal growth factor (EGF; Biosource, Karlsruhe, Germany), 20 ng/mL recombinant human fibroblast growth factor (rhFGF; R&D Systems, Wiesbaden-Nordenstadt, Germany), and penicillin and streptomycin (1:100; Invitrogen) at 37°C with 7.5% CO2 . Differentiation was initiated by growth factor withdrawal and plating onto a poly-d-lysine/ laminin matrix. The neurospheres were fed every 2-3 days by replacing 50% of the culture media with fresh media. The culture suspension was transferred to a dish containing 16 wells. The wells were divided as follows: 4 wells received no papaverine (control), 4 wells 1 u, 4 wells 5 u and 4 wells 10 u of papaverine solution. In the next 2 weeks, photography (0,4,5,11days) and viability test were done. The photographs were analysed. Results : papaverine didn't affect proliferation of neurospheres, while it affected viability compared to control , this was dose related. Conclusion: This indicates the harmful effect of papaverine suggesting it to be a candidate neurotoxin causing Parkinsonism.

Keywords: neurospheres, neural stem cells, papaverine, Parkinsonism

Procedia PDF Downloads 660

Authors: Prikhil Agrawal, N. Priyanka

Abstract:

With the increase of millions of users on Internet day by day, it is very essential to maintain highly reliable and secured data communication between various corporations. Although there are various traditional security imparting techniques such as antivirus software, password protection, data encryption, biometrics and firewall etc. But still network security has become the main issue in various leading companies. So IDSs have become an essential component in terms of security, as it can detect various network attacks and respond quickly to such occurrences. IDSs are used to detect unauthorized access to a computer system. This paper describes various intrusion detection techniques using GA approach. The intrusion detection problem has become a challenging task due to the conception of miscellaneous computer networks under various vulnerabilities. Thus the damage caused to various organizations by malicious intrusions can be mitigated and even be deterred by using this powerful tool.

Keywords: genetic algorithm (GA), intrusion detection system (IDS), dataset, network security

Procedia PDF Downloads 297

Authors: S. M. Gupta, Vanita Aggarwal, Som Nath Sachdeva

Abstract:

The work reported through this paper is an experimental work conducted on High Performance Concrete (HPC) with super plasticizer with the aim to develop some models suitable for prediction of compressive strength of HPC mixes. In this study, the effect of varying proportions of fly ash (0% to 50% at 10% increment) on compressive strength of high performance concrete has been evaluated. The mix designs studied were M30, M40 and M50 to compare the effect of fly ash addition on the properties of these concrete mixes. In all eighteen concrete mixes have been designed, three as conventional concretes for three grades under discussion and fifteen as HPC with fly ash with varying percentages of fly ash. The concrete mix designing has been done in accordance with Indian standard recommended guidelines i.e. IS: 10262. All the concrete mixes have been studied in terms of compressive strength at 7 days, 28 days, 90 days and 365 days. All the materials used have been kept same throughout the study to get a perfect comparison of values of results. The models for compressive strength prediction have been developed using Linear Regression method (LR), Artificial Neural Network (ANN) and Leave One Out Validation (LOOV) methods.

Keywords: high performance concrete, fly ash, concrete mixes, compressive strength, strength prediction models, linear regression, ANN

Procedia PDF Downloads 444

Authors: Qitao Xie, Qingquan Zhang, Xiaofei Zhang, Di Tian, Ruixuan Wen, Ting Zhu, Ping Yi, Xin Li

Abstract:

Inspired by the recent movement of digital currency, we are building a question answering system concerning the subject of cryptocurrency using Bidirectional Encoder Representations from Transformers (BERT). The motivation behind this work is to properly assist digital currency investors by directing them to the corresponding knowledge bases that can offer them help and increase the querying speed. BERT, one of newest language models in natural language processing, was investigated to improve the quality of generated responses. We studied different combinations of hyperparameters of the BERT model to obtain the best fit responses. Further, we created an intelligent chatbot for cryptocurrency using BERT. A chatbot using BERT shows great potential for the further advancement of a cryptocurrency market tool. We show that the BERT neural networks generalize well to other tasks by applying it successfully to cryptocurrency.

Keywords: bidirectional encoder representations from transformers, BERT, chatbot, cryptocurrency, deep learning

Procedia PDF Downloads 147

Authors: Nadia el Atlas, Mohammed el Aroussi, Mohammed Wahbi

Abstract:

In this paper a novel technique of mass characterization based on robust features-fusion is presented. The proposed method consists of mainly four stages: (a) the first phase involves segmenting the masses using edge information’s. (b) The second phase is to calculate and fuse the most relevant morphological features. (c) The last phase is the classification step which allows us to classify the images into benign and malignant masses. In this step we have implemented Support Vectors Machines (SVM) and Artificial Neural Networks (ANN), which were evaluated with the following performance criteria: confusion matrix, accuracy, sensitivity, specificity, receiver operating characteristic ROC, and error histogram. The effectiveness of this new approach was evaluated by a recently developed database: INBREAST database. The fusion of the most appropriate morphological features provided very good results. The SVM gives accuracy to within 64.3%. Whereas the ANN classifier gives better results with an accuracy of 97.5%.

Keywords: breast cancer, mammography, CAD system, features, fusion

Procedia PDF Downloads 599

Authors: Kavyadevi N., Kaviya G., Gowsalya P., Janani M., Mohanraj S.

Abstract:

Hevea brasiliensis, also known as the rubber tree, is one of the foremost assets of crops in the world. One of the most significant advantages of the Rubber Plant in terms of air oxygenation is its capacity to reduce the likelihood of an individual developing respiratory allergies like asthma. To construct such a system that can properly identify crop diseases and pests and then create a database of insecticides for each pest and disease, we must first give treatment for the illness that has been detected. We shall primarily examine three major leaf diseases since they are economically deficient in this article, which is Bird's eye spot, algal spot and powdery mildew. And the recommended work focuses on disease identification on rubber tree leaves. It will be accomplished by employing one of the superior algorithms. Input, Preprocessing, Image Segmentation, Extraction Feature, and Classification will be followed by the processing technique. We will use time-consuming procedures that they use to detect the sickness. As a consequence, the main ailments, underlying causes, and signs and symptoms of diseases that harm the rubber tree are covered in this study.

Keywords: image processing, python, convolution neural network (CNN), machine learning

Procedia PDF Downloads 76

Authors: Mohd Zamri Osman, Mohd Aizaini Maarof, Mohd Foad Rohani

Abstract:

Human skin detection recognized as the primary step in most of the applications such as face detection, illicit image filtering, hand recognition and video surveillance. The performance of any skin detection applications greatly relies on the two components: feature extraction and classification method. Skin color is the most vital information used for skin detection purpose. However, color feature alone sometimes could not handle images with having same color distribution with skin color. A color feature of pixel-based does not eliminate the skin-like color due to the intensity of skin and skin-like color fall under the same distribution. Hence, the statistical color analysis will be exploited such mean and standard deviation as an additional feature to increase the reliability of skin detector. In this paper, we studied the effectiveness of statistical color feature for human skin detection. Furthermore, the paper analyzed the integrated color and texture using eight classifiers with three color spaces of RGB, YCbCr, and HSV. The experimental results show that the integrating statistical feature using Random Forest classifier achieved a significant performance with an F1-score 0.969.

Keywords: color space, neural network, random forest, skin detection, statistical feature

Procedia PDF Downloads 462

Authors: Z. Gharaee

Abstract:

Recognizing actions performed by others is important in our daily lives since it is necessary for communicating with others in a proper way. We perceive an action by observing the kinematics of motions involved in the performance. We use our experience and concepts to make a correct recognition of the actions. Although building the action concepts is a life-long process, which is repeated throughout life, we are very efficient in applying our learned concepts in analyzing motions and recognizing actions. Experiments on the subjects observing the actions performed by an actor show that an action is recognized after only about two hundred milliseconds of observation. In this study, hierarchical action recognition architecture is proposed by using growing grid layers. The first-layer growing grid receives the pre-processed data of consecutive 3D postures of joint positions and applies some heuristics during the growth phase to allocate areas of the map by inserting new neurons. As a result of training the first-layer growing grid, action pattern vectors are generated by connecting the elicited activations of the learned map. The ordered vector representation layer receives action pattern vectors to create time-invariant vectors of key elicited activations. Time-invariant vectors are sent to second-layer growing grid for categorization. This grid creates the clusters representing the actions. Finally, one-layer neural network developed by a delta rule labels the action categories in the last layer. System performance has been evaluated in an experiment with the publicly available MSR-Action3D dataset. There are actions performed by using different parts of human body: Hand Clap, Two Hands Wave, Side Boxing, Bend, Forward Kick, Side Kick, Jogging, Tennis Serve, Golf Swing, Pick Up and Throw. The growing grid architecture was trained by applying several random selections of generalization test data fed to the system during on average 100 epochs for each training of the first-layer growing grid and around 75 epochs for each training of the second-layer growing grid. The average generalization test accuracy is 92.6%. A comparison analysis between the performance of growing grid architecture and self-organizing map (SOM) architecture in terms of accuracy and learning speed show that the growing grid architecture is superior to the SOM architecture in action recognition task. The SOM architecture completes learning the same dataset of actions in around 150 epochs for each training of the first-layer SOM while it takes 1200 epochs for each training of the second-layer SOM and it achieves the average recognition accuracy of 90% for generalization test data. In summary, using the growing grid network preserves the fundamental features of SOMs, such as topographic organization of neurons, lateral interactions, the abilities of unsupervised learning and representing high dimensional input space in the lower dimensional maps. The architecture also benefits from an automatic size setting mechanism resulting in higher flexibility and robustness. Moreover, by utilizing growing grids the system automatically obtains a prior knowledge of input space during the growth phase and applies this information to expand the map by inserting new neurons wherever there is high representational demand.

Keywords: action recognition, growing grid, hierarchical architecture, neural networks, system performance

Procedia PDF Downloads 157

Authors: Azeddine El-Hassouny, Chandrashekhar Meshram, Geraldin Nanfack

Abstract:

In recent years, learning from data with noisy labels (Label Noise) has been a major concern in supervised learning. This problem has become even more worrying in Deep Learning, where the generalization capabilities have been questioned lately. Indeed, deep learning requires a large amount of data that is generally collected by search engines, which frequently return data with unreliable labels. In this paper, we investigate the Label Noise in Deep Learning using variational inference. Our contributions are : (1) exploiting Label Noise concept where the true labels are learnt using reparameterization variational inference, while observed labels are learnt discriminatively. (2) the noise transition matrix is learnt during the training without any particular process, neither heuristic nor preliminary phases. The theoretical results shows how true label distribution can be learned by variational inference in any discriminate neural network, and the effectiveness of our approach is proved in several target datasets, such as MNIST and CIFAR32.

Keywords: label noise, deep learning, discrete latent variable, variational inference, MNIST, CIFAR32

Procedia PDF Downloads 128

Authors: Heba M. Wagih, Hoda M. O. Mokhtar

Abstract:

Social Network has recently played a significant role in both scientific and social communities. The growing adoption of social network applications has been a relevant source of information nowadays. Due to its popularity, several research trends are emerged to service the huge volume of users including, Location-Based Social Networks (LBSN), Recommendation Systems, Sentiment Analysis Applications, and many others. LBSNs applications are among the highly demanded applications that do not focus only on analyzing the spatiotemporal positions in a given raw trajectory but also on understanding the semantics behind the dynamics of the moving object. LBSNs are possible means of predicting human mobility based on users social ties as well as their spatial preferences. LBSNs rely on the efficient representation of users’ trajectories. Hence, traditional raw trajectory information is no longer convenient. In our research, we focus on studying human behavior trajectory which is the major pillar in location recommendation systems. In this paper, we propose an ontology design patterns with their underlying description logics to efficiently annotate human behavior trajectories.

Keywords: human behavior trajectory, location-based social network, ontology, social network

Procedia PDF Downloads 452

Authors: Hossam Abdelrahman, Sama Rostom, Reem Yassein, Yara Mohamed, Salma Salah, Nour Awny

Abstract:

In today’s environment, people are becoming increasingly interested in their appearance. However, they are afraid of their unknown appearance after a plastic surgery or treatment. Accidents, burns and genetic problems such as bowing of body parts of people have a negative impact on their mental health with their appearance and this makes them feel uncomfortable and underestimated. The approach presents a revolutionary deep learning-based image inpainting method that analyses the various picture structures and corrects damaged images. In this study, A model is proposed based on the in-painting of medical images with Stable Diffusion Inpainting method. Reconstructing missing and damaged sections of an image is known as image inpainting is a key progress facilitated by deep neural networks. The system uses the input of the user of an image to indicate a problem, the system will then modify the image and output the fixed image, facilitating for the patient to see the final result.

Keywords: generative adversarial network, large mask inpainting, stable diffusion inpainting, plastic surgery

Procedia PDF Downloads 74

Authors: Krishan P. Sharma, T. P. Sharma

Abstract:

Wireless sensor nodes are resource constrained battery powered devices usually deployed in hostile and ill-disposed areas to cooperatively monitor physical or environmental conditions. Due to their limited power supply, the major challenge for researchers is to utilize their battery power for enhancing the lifetime of whole network. Communication and sensing are two major sources of energy consumption in sensor networks. In this paper, we propose a deployment strategy for enhancing the average lifetime of a sensor network by effectively utilizing communication and sensing energy to provide full coverage. The proposed scheme is based on the fact that due to heavy relaying load, sensor nodes near to the sink drain energy at much faster rate than other nodes in the network and consequently die much earlier. To cover this imbalance, proposed scheme finds optimal communication and sensing ranges according to effective load at each node and uses a non-uniform deployment strategy where there is a comparatively high density of nodes near to the sink. Probable relaying load factor at particular node is calculated and accordingly optimal communication distance and sensing range for each sensor node is adjusted. Thus, sensor nodes are placed at locations that optimize energy during network operation. Formal mathematical analysis for calculating optimized locations is reported in present work.

Keywords: load factor, network lifetime, non-uniform deployment, sensing range

Procedia PDF Downloads 383