Search results for: convolutional networks

Authors: Deokho Kim, Wonwoo Lee, Jaewoong Lee, Teresa Ng, Gun-Ill Lee, Jiwon Jeong

Abstract:

This paper presents a method for improving object search accuracy using a deep learning model. A major limitation to provide accurate similarity with deep learning is the requirement of huge amount of data for training pairwise similarity scores (metrics), which is impractical to collect. Thus, similarity scores are usually trained with a relatively small dataset, which comes from a different domain, causing limited accuracy on measuring similarity. For this reason, this paper proposes a deep learning model that can be trained with a significantly small amount of data, a clustered data which of each cluster contains a set of visually similar images. In order to measure similarity distance with the proposed method, visual features of two images are extracted from intermediate layers of a convolutional neural network with various pooling methods, and the network is trained with pairwise similarity scores which is defined zero for images in identical cluster. The proposed method outperforms the state-of-the-art object similarity scoring techniques on evaluation for finding exact items. The proposed method achieves 86.5% of accuracy compared to the accuracy of the state-of-the-art technique, which is 59.9%. That is, an exact item can be found among four retrieved images with an accuracy of 86.5%, and the rest can possibly be similar products more than the accuracy. Therefore, the proposed method can greatly reduce the amount of training data with an order of magnitude as well as providing a reliable similarity metric.

Keywords: visual search, deep learning, convolutional neural network, machine learning

Procedia PDF Downloads 194

Authors: Oluwatosin Olofintoye, Josiah Adeyemo, Gbemileke Shomade

Abstract:

The relationship between rainfall and runoff is an important issue in surface water hydrology therefore the understanding and development of accurate rainfall-runoff models and their applications in water resources planning, management and operation are of paramount importance in hydrological studies. This paper reviews some of the previous works on the rainfall-runoff process modeling. The hydrologic applications of conceptual models and artificial neural networks (ANNs) for the precipitation-runoff process modeling were studied. Gradient training methods such as error back-propagation (BP) and evolutionary algorithms (EAs) are discussed in relation to the training of artificial neural networks and it is shown that application of EAs to artificial neural networks training could be an alternative to other training methods. Therefore, further research interest to exploit the abundant expert knowledge in the area of artificial intelligence for the solution of hydrologic and water resources planning and management problems is needed.

Keywords: artificial intelligence, artificial neural networks, evolutionary algorithms, gradient training method, rainfall-runoff model

Procedia PDF Downloads 422

Authors: Ibrahim Obeidat

Abstract:

Reaching a fully-connected network of mobile nodes in rural areas got a great attention between network researchers. This attention rose due to the complexity and high costs while setting up the needed infrastructures for these networks, in addition to the low transmission range these nodes has. Terranet technology, as an example, employs ad-hoc mesh network where each node has a transmission range not exceed one kilometer, this means that every two nodes are able to communicate with each other if they are just one kilometer far from each other, otherwise a third-party will play the role of the “relay”. In Terranet, and as an idea to reduce network setup cost, every node in the network will be considered as a router that is responsible of forwarding data between other nodes which result in a decentralized collaborative environment. Most researches on Terranet presents the idea of how to encourage mobile nodes to become more cooperative by letting their devices in “ON” state as long as possible while accepting to play the role of relay (router). This research presents the issue of finding the percentage of nodes in ad-hoc mesh network within rural areas that should play the role of relay at every time slot, relating to what is the actual area coverage of nodes in order to have the network reach the fully-connectivity. Far from our knowledge, till now there is no current researches discussed this issue. The research is done by making an implementation that depends on building adjacency matrix as an indicator to the connectivity between network members. This matrix is continually updated until each value in it refers to the number of hubs that should be followed to reach from one node to another. After repeating the algorithm on different area sizes, different coverage percentages for each size, and different relay percentages for several times, results extracted shows that for area coverage less than 5% we need to have 40% of the nodes to be relays, where 10% percentage is enough for areas with node coverage greater than 5%.

Keywords: ad-hoc mesh networks, network connectivity, mobile ad-hoc networks, Terranet, adjacency matrix, simulator, wireless sensor networks, peer to peer networks, vehicular Ad hoc networks, relay

Procedia PDF Downloads 252

Authors: Asma Ben Yaghlane, Mohamed Naceur Azaiez, Mehdi Mrad

Abstract:

We investigate the large scale of networks in the context of network survivability under attack. We use appropriate techniques to evaluate and the attacker-based- and the defender-based-network survivability. The attacker is unaware of the operated links by the defender. Each attacked link has some pre-specified probability to be disconnected. The defender choice is so that to maximize the chance of successfully sending the flow to the destination node. The attacker however will select the cut-set with the highest chance to be disabled in order to partition the network. Moreover, we extend the problem to the case of selecting the best p paths to operate by the defender and the best k cut-sets to target by the attacker, for arbitrary integers p,k > 1. We investigate some variations of the problem and suggest polynomial-time solutions.

Keywords: defense/attack strategies, large scale, networks, partitioning a network

Procedia PDF Downloads 259

Authors: Babak Mohajeri

Abstract:

In recent years, the conventional business model of ownership has been changed towards accessibility in a variety of markets. Two trends can be observed in the evolution of this rental-like business model. Firstly, the technological development that enables the emergence of new business models. These new business models increasingly become agile and flexible. For example Spotify, an online music stream company provides consumers access to over millions of music tracks, conveniently through the smartphone, tablet or computer. Similarly, Car2Go, the car sharing company accesses its members with flexible and nearby sharing cars. The second trend is the increasing communication and connections via social networks. This trend enables a shift to peer-to-peer accessibility based business models. Conventionally, companies provide access for their customers to own companies products or services. In peer-to-peer model, nonetheless, companies facilitate access and connection across their customers to use other customers owned property or skills, competencies or services .The is so-called the sharing economy business model. The aim of this study is to investigate into a new and emerging type of the sharing economy model in which role of customers and service providers may dramatically change. This new model is called Collaborative Service Networks. We propose a mechanism for Collaborative Service Networks business model. Uber and Airbnb, two successful growing companies, have been selected for our case studies and their business models are analyzed. Finally, we study the emergence of the collaborative service networks in the manufacturing domain. Our finding results to a new manufacturing paradigm called social manufacturing.

Keywords: sharing economy, collaborative service networks, social manufacturing, manufacturing development

Procedia PDF Downloads 295

Authors: Wheyming Tina Song, Chi-Hao Hong, Peisyuan Lin

Abstract:

An important performance measure for stochastic manufacturing networks is the system reliability, defined as the probability that the production output meets or exceeds a specified demand. The system parameters include the capacity of each workstation and numbers of the conforming parts produced in each workstation. We establish that eighteen archival publications, containing twenty-one examples, provide incorrect values of the system reliability. The author recently published the Song Rule, which provides the correct analytical system-reliability value; it is, however, computationally inefficient for large networks. In this paper, we use Monte Carlo simulation (implemented in C and Flexsim) to provide estimates for the above-mentioned twenty-one examples. The simulation estimates are consistent with the analytical solution for small networks but is computationally efficient for large networks. We argue here for three advantages of Monte Carlo simulation: (1) understanding stochastic systems, (2) validating analytical results, and (3) providing estimates even when analytical and numerical approaches are overly expensive in computation. Monte Carlo simulation could have detected the published analysis errors.

Keywords: Monte Carlo simulation, analytical results, leading digit rule, standard error

Procedia PDF Downloads 337

Authors: N. Nalini, Lokesh B. Bhajantri

Abstract:

In Distributed Sensor Networks, the sensor nodes are prone to failure due to energy depletion and some other reasons. In this regard, fault tolerance of network is essential in distributed sensor environment. Energy efficiency, network or topology control and fault-tolerance are the most important issues in the development of next-generation Distributed Sensor Networks (DSNs). This paper proposes a node fault detection and recovery using Genetic Algorithm (GA) in DSN when some of the sensor nodes are faulty. The main objective of this work is to provide fault tolerance mechanism which is energy efficient and responsive to network using GA, which is used to detect the faulty nodes in the network based on the energy depletion of node and link failure between nodes. The proposed fault detection model is used to detect faults at node level and network level faults (link failure and packet error). Finally, the performance parameters for the proposed scheme are evaluated.

Keywords: distributed sensor networks, genetic algorithm, fault detection and recovery, information technology

Procedia PDF Downloads 428

Authors: Ramazan I. Kadiev, Arcady Ponosov

Abstract:

Mathematical models of gene regulatory networks can in many cases be described by ordinary differential equations with switching nonlinearities, where the initial value problem is ill-posed. Several regularization methods are known in the case of deterministic networks, but the presence of stochastic noise leads to several technical difficulties. In the presentation, it is proposed to apply the methods of the stochastic singular perturbation theory going back to Yu. Kabanov and Yu. Pergamentshchikov. This approach is used to regularize the above ill-posed problem, which, e.g., makes it possible to design stable numerical schemes. Several examples are provided in the presentation, which support the efficiency of the suggested analysis. The method can also be of interest in other fields of biomathematics, where differential equations contain switchings, e.g., in neural field models.

Keywords: ill-posed problems, singular perturbation analysis, stochastic differential equations, switching nonlinearities

Procedia PDF Downloads 174

Authors: L. Parisi

Abstract:

Kinematic data wisely correlate vector quantities in space to scalar parameters in time to assess the degree of symmetry between the intact limb and the amputated limb with respect to a normal model derived from the gait of control group participants. Furthermore, these particular data allow a doctor to preliminarily evaluate the usefulness of a certain rehabilitation therapy. Kinetic curves allow the analysis of ground reaction forces (GRFs) to assess the appropriateness of human motion. Electromyography (EMG) allows the analysis of the fundamental lower limb force contributions to quantify the level of gait asymmetry. However, the use of this technological tool is expensive and requires patient’s hospitalization. This research work suggests overcoming the above limitations by applying artificial neural networks.

Keywords: kinetics, kinematics, cyclograms, neural networks, transtibial amputation

Procedia PDF Downloads 424

Authors: Asma Ben Yaghlane, Mohamed Naceur Azaiez

Abstract:

We consider the problem of attacks on networks. We define the concept of system survivability in networks in the presence of intelligent threats. Our setting of the problem assumes a flow to be sent from one source node to a destination node. The attacker attempts to disable the network by preventing the flow to reach its destination while the defender attempts to identify the best path-set to use to maximize the chance of arrival of the flow to the destination node. Our concept is shown to be different from the classical concept of network reliability. We distinguish two types of network survivability related to the defender and to the attacker of the network, respectively. We prove that the defender-based-network survivability plays the role of a lower bound while the attacker-based-network survivability plays the role of an upper bound of network reliability. We also prove that both concepts almost never agree nor coincide with network reliability. Moreover, we use the shortest-path problem to determine the defender-based-network survivability and the min-cut problem to determine the attacker-based-network survivability. We extend the problem to a variety of models including the minimum-spanning-tree problem and the multiple source-/destination-network problems.

Keywords: defense/attack strategies, information, networks, reliability, survivability

Procedia PDF Downloads 364

Authors: Mehrshad Khosraviani, Faramarz Abbaspour Leyl Abadi

Abstract:

The fundamental component of the computer network of modern information society will be considered. These networks are connected to the network of the internet generally. Due to the fact that the primary purpose of the Internet is not designed for, in recent decades, none of these networks in many of the attacks has been very important. Today, for the provision of security, different security tools and systems, including intrusion detection systems are used in the network. A common diagnosis system based on artificial immunity, the designer, the Adhasaz Foundation has been evaluated. The idea of using artificial safety methods in the diagnosis of abnormalities in computer networks it has been stimulated in the direction of their specificity, there are safety systems are similar to the common needs of m, that is non-diagnostic. For example, such methods can be used to detect any abnormalities, a variety of attacks, being memory, learning ability, and Khodtnzimi method of artificial immune algorithm pointed out. Diagnosis of the common system of education offered in this paper using only the normal samples is required for network and any additional data about the type of attacks is not. In the proposed system of positive selection and negative selection processes, selection of samples to create a distinction between the colony of normal attack is used. Copa real data collection on the evaluation of ij indicates the proposed system in the false alarm rate is often low compared to other ir methods and the detection rate is in the variations.

Keywords: artificial immune system, abnormality detection, intrusion detection, computer networks

Procedia PDF Downloads 332

Authors: R. Schweighoffer, N. Nagy, E. Reeves, B. Liebig

Abstract:

Due to aging populations, the need for seamless palliative care provision is of central interest for western societies. An essential aspect of palliative care delivery is the quality of collaboration amongst palliative care providers. Therefore, the current research is based on Bainbridge’s conceptual framework, which provides an outline for the evaluation of palliative care provision. This study is the first one to investigate the predictive validity of spatial distribution on the quantity of interaction amongst various palliative care providers. Furthermore, based on the familiarity principle, we examine whether the extent of collaboration influences the perceived quality of collaboration among palliative care providers in urban versus rural areas of Switzerland. Based on a population-representative survey of Swiss palliative care providers, the results of the current study show that professionals in densely populated areas report higher absolute numbers of interactions and are more satisfied with their collaborative practice. This indicates that palliative care providers who work in urban areas are better embedded into networks than their counterparts in more rural areas. The findings are especially important, considering that efficient collaboration is a prerequisite to achieve satisfactory patient outcomes. Conclusively, measures should be taken to foster collaboration in weakly interconnected palliative care networks.

Keywords: collaboration, healthcare networks, palliative care, Switzerland

Procedia PDF Downloads 240

Authors: Masood Varshosaz, Kamyar Hasanpour

Abstract:

In recent years, we have seen growing interest in recognizing humans in drone images for post-disaster search and rescue operations. Deep learning algorithms have shown great promise in this area, but they often require large amounts of labeled data to train the models. To keep the data acquisition cost low, augmentation techniques can be used to create additional data from existing images. There are many techniques of such that can help generate variations of an original image to improve the performance of deep learning algorithms. While data augmentation is potentially assumed to improve the accuracy and robustness of the models, it is important to ensure that the performance gains are not outweighed by the additional computational cost or complexity of implementing the techniques. To this end, it is important to evaluate the impact of data augmentation on the performance of the deep learning models. In this paper, we evaluated the most currently available 2D data augmentation techniques on a standard convolutional network which was trained for recognizing humans in drone images. The techniques include rotation, scaling, random cropping, flipping, shifting, and their combination. The results showed that the augmented models perform 1-3% better compared to a base network. However, as the augmented images only contain the human parts already visible in the original images, a new data augmentation approach is needed to include the invisible parts of the human body. Thus, we suggest a new method that employs simulated 3D human models to generate new data for training the network.

Keywords: human recognition, deep learning, drones, disaster mitigation

Procedia PDF Downloads 71

Authors: Thiago Spilborghs Bueno Meyer, Plinio Thomaz Aquino Junior

Abstract:

Through speech, which privileges the functional and interactive nature of the text, it is possible to ascertain the spatiotemporal circumstances, the conditions of production and reception of the discourse, the explicit purposes such as informing, explaining, convincing, etc. These conditions allow bringing the interaction between humans closer to the human-robot interaction, making it natural and sensitive to information. However, it is not enough to understand what is said; it is necessary to recognize emotions for the desired interaction. The validity of the use of neural networks for feature selection and emotion recognition was verified. For this purpose, it is proposed the use of neural networks and comparison of models, such as recurrent neural networks and deep neural networks, in order to carry out the classification of emotions through speech signals to verify the quality of recognition. It is expected to enable the implementation of robots in a domestic environment, such as the HERA robot from the RoboFEI@Home team, which focuses on autonomous service robots for the domestic environment. Tests were performed using only the Mel-Frequency Cepstral Coefficients, as well as tests with several characteristics of Delta-MFCC, spectral contrast, and the Mel spectrogram. To carry out the training, validation and testing of the neural networks, the eNTERFACE’05 database was used, which has 42 speakers from 14 different nationalities speaking the English language. The data from the chosen database are videos that, for use in neural networks, were converted into audios. It was found as a result, a classification of 51,969% of correct answers when using the deep neural network, when the use of the recurrent neural network was verified, with the classification with accuracy equal to 44.09%. The results are more accurate when only the Mel-Frequency Cepstral Coefficients are used for the classification, using the classifier with the deep neural network, and in only one case, it is possible to observe a greater accuracy by the recurrent neural network, which occurs in the use of various features and setting 73 for batch size and 100 training epochs.

Keywords: emotion recognition, speech, deep learning, human-robot interaction, neural networks

Procedia PDF Downloads 136

Authors: A. Degale Desta, Cheng Jian

Abstract:

Deep learning application in computer vision is rapidly advancing, giving it the ability to monitor the public and quickly identify potentially anomalous behaviour from crowd scenes. Therefore, the purpose of the current work is to improve the performance of safety of people in crowd events from panic behaviour through introducing the innovative idea of Aggregation of Ensembles (AOE), which makes use of the pre-trained ConvNets and a pool of classifiers to find anomalies in video data with packed scenes. According to the theory of algorithms that applied K-means, KNN, CNN, SVD, and Faster-CNN, YOLOv5 architectures learn different levels of semantic representation from crowd videos; the proposed approach leverages an ensemble of various fine-tuned convolutional neural networks (CNN), allowing for the extraction of enriched feature sets. In addition to the above algorithms, a long short-term memory neural network to forecast future feature values and a handmade feature that takes into consideration the peculiarities of the crowd to understand human behavior. On well-known datasets of panic situations, experiments are run to assess the effectiveness and precision of the suggested method. Results reveal that, compared to state-of-the-art methodologies, the system produces better and more promising results in terms of accuracy and processing speed.

Keywords: action recognition, computer vision, crowd detecting and tracking, deep learning

Procedia PDF Downloads 134

Authors: Ashish Jaiswal, Ashwin Ramesh Babu, Mohammad Zaki Zadeh, Fillia Makedon, Glenn Wylie

Abstract:

Functional magnetic resonance imaging (fMRI) is a neuroimaging technique that records neural activations in the brain by capturing the blood oxygen level in different regions based on the task performed by a subject. Given fMRI data, the problem of predicting the state of cognitive fatigue in a person has not been investigated to its full extent. This paper proposes tackling this issue as a multi-class classification problem by dividing the state of cognitive fatigue into six different levels, ranging from no-fatigue to extreme fatigue conditions. We built a spatio-temporal model that uses convolutional neural networks (CNN) for spatial feature extraction and a long short-term memory (LSTM) network for temporal modeling of 4D fMRI scans. We also applied a self-supervised method called MoCo (Momentum Contrast) to pre-train our model on a public dataset BOLD5000 and fine-tuned it on our labeled dataset to predict cognitive fatigue. Our novel dataset contains fMRI scans from Traumatic Brain Injury (TBI) patients and healthy controls (HCs) while performing a series of N-back cognitive tasks. This method establishes a state-of-the-art technique to analyze cognitive fatigue from fMRI data and beats previous approaches to solve this problem.

Keywords: fMRI, brain imaging, deep learning, self-supervised learning, contrastive learning, cognitive fatigue

Procedia PDF Downloads 164

Authors: Shahed Vahedi, Mohd Salmi Md Noorani

Abstract:

This paper investigates cluster synchronization phenomena between community networks. We focus on the situation where a variety of dynamics occur in the clusters. In particular, we show that different synchronization states simultaneously occur between the networks. The controller is designed having an adaptive control gain, and theoretical results are derived via Lyapunov stability. Simulations on well-known dynamical systems are provided to elucidate our results.

Keywords: cluster synchronization, adaptive control, community network, simulation

Procedia PDF Downloads 453

Authors: Nada Souissi, Mourad Mroua

Abstract:

The main purpose of this study is to examine the interaction between financial asset volatility, economic factors and investor's behavioral indicators related to both the company's and the markets stocks for the period from January 2000 to January2020. Using multiple linear regression and Bayesian Networks modeling, results show a positive and negative relationship between investor's psychology index, economic factors and predicted stock market return. We reveal that the application of the Bayesian Discrete Network contributes to identify the different cause and effect relationships between all economic, financial variables and psychology index.

Keywords: Financial asset return predictability, Economic factors, Investor's psychology index, Bayesian approach, Probabilistic networks, Parametric learning

Procedia PDF Downloads 118

Authors: Wen-Bin Liau, Wan-Ting Wang, Chiang-Jen Hsiao, Sheng-Mao Tseng

Abstract:

In this paper, an interesting and easy method to prepare one-dimensional nanostructured polyaniline/dendritic silver composites is reported. It is well known that the morphology of metal particle is a very important factor to influence the properties of polymer-metal composites. Usually, the dendritic silver is prepared by kinetic control in reduction reaction. It is not a thermodynamically stable structure. It is the goal to reduce silver ion to dendritic silver by polyaniline polymer via kinetic control and form one-dimensional nanostructured polyaniline/dendritic silver composites. The preparation is a two steps sequential reaction. First step, the polyaniline networks composed of nano fibrillar polyaniline are synthesized from aniline monomers aqueous with ammonium persulfate as the initiator at room temperature. In second step, the silver nitrate is added into polyaniline networks dispersed in deionized water. The dendritic silver is formed via reduction by polyaniline networks under the kinetic control. The formation of polyaniline is discussed via transmission electron microscopy (TEM). Nanosheets, nanotubes, nanospheres, nanosticks, and networks are observed via TEM. Then, the mechanism of formation of one-dimensional nanostructured polyaniline/dendritic silver composites is discussed. The formation of dendritic silver is observed by TEM and X-ray diffraction.

Keywords: 1D nanostructured polyaniline, dendritic silver, synthesis

Procedia PDF Downloads 476

Authors: Farouk Shehu Abdulwahab

Abstract:

The concept of farmer-farmer knowledge sharing was introduced to bridge the extension worker-farmer ratio gap in developing countries. However, the idea was poorly accepted, especially in typical agrarian communities. Therefore, the study explores the concept of a farmer-to-farmer knowledge-sharing network to enhance extension service delivery. The study collected data from 80 farmers randomly selected through a series of multiple stages. The Data was analysed using a 5-point Likert scale and descriptive statistics. The Likert scale results revealed that 62.5% of the farmers are satisfied with farmer-to-farmer knowledge-sharing networks. Moreover, descriptive statistics show that lack of capacity building and low level of education are the most significant problems affecting farmer-farmer sharing networks. The major implication of these findings is that the concept of farmer-farmer knowledge-sharing networks can work better for farmers in developing countries as it was perceived by them as a reliable alternative for information sharing. Therefore, the study recommends introducing incentives into the concept of farmer-farmer knowledge-sharing networks and enhancing the capabilities of farmers who are opinion leaders in the farmer-farmer concept of knowledge-sharing to make it more sustainable.

Keywords: agricultural productivity, extension, farmer-to-farmer, livelihood, technology transfer

Procedia PDF Downloads 38

Authors: Derya Nizam

Abstract:

An alternative understanding of "localization" has gained significance as the ecological and social issues associated with the growing pressure of agricultural homogeneity and standardization become more apparent. Through an analysis of a case study on an alternative food networks in Turkey, this research seeks to critically examine the localization movement. The results indicate that the idea of localization helps to create new niche markets by creating place-based labels, but it also strengthens local identities through social networks that connect rural and urban areas. In that context, localization manifests as a commodification movement that appropriates local and cultural values to generate capitalist profit, as well as a grassroots movement that strengthens the resilience of local communities. This research addresses the potential of community development approaches in the democratization of global agro-food networks.

Keywords: community building, local food, alternative food movements, localization

Procedia PDF Downloads 52

Authors: Krishna Mohan Bathula, Fatou Bintou Loucoubar, FNU Kaleemunnisa, Christelle Scharff, Mark Anthony De Castro

Abstract:

Voice recognition algorithms such as automatic speech recognition and text-to-speech systems with African languages can play an important role in bridging the digital divide of Artificial Intelligence in Africa, contributing to the establishment of a fully inclusive information society. This paper proposes a Deep Learning model that can classify the user responses as inputs for an interactive voice response system. A dataset with Wolof language words ‘yes’ and ‘no’ is collected as audio recordings. A two stage Data Augmentation approach is adopted for enhancing the dataset size required by the deep neural network. Data preprocessing and feature engineering with Mel-Frequency Cepstral Coefficients are implemented. Convolutional Neural Networks (CNNs) have proven to be very powerful in image classification and are promising for audio processing when sounds are transformed into spectra. For performing voice response classification, the recordings are transformed into sound frequency feature spectra and then applied image classification methodology using a deep CNN model. The inference model of this trained and reusable Wolof voice response recognition system can be integrated with many applications associated with both web and mobile platforms.

Keywords: automatic speech recognition, interactive voice response, voice response recognition, wolof word classification

Procedia PDF Downloads 91

Authors: Vaidehi Iyer, Konstantin Borozdin

Abstract:

Time series prediction problems have many important practical applications, but are notoriously difficult for statistical modeling. Recently, machine learning methods have been attracted significant interest as a practical tool applied to a variety of problems, even though developments in this field tend to be semi-empirical. This paper explores application of Long Short Term Memory based Recurrent Neural Networks to the one-step prediction of time series for both trend and stochastic components. Two types of data are analyzed - daily stock prices, that are often considered to be a typical example of a random walk, - and weather patterns dominated by seasonal variations. Results from both analyses are compared, and reinforced learning framework is used to select more efficient between Recurrent Neural Networks and more traditional auto regression methods. It is shown that both methods are able to follow long-term trends and seasonal variations closely, but have difficulties with reproducing day-to-day variability. Future research directions and potential real world applications are briefly discussed.

Keywords: long short term memory, prediction methods, recurrent neural networks, reinforcement learning

Procedia PDF Downloads 206

Authors: Erick C. Valverde, Wansu Lim

Abstract:

Facial emotion recognition (FER) system has been recently developed for more advanced computer vision applications. The ability to identify human emotions would enable smart healthcare facility to diagnose mental health illnesses (e.g., depression and stress) as well as better human social interactions with smart technologies. The FER system involves two steps: 1) face detection task and 2) facial emotion recognition task. It classifies the human expression in various categories such as angry, disgust, fear, happy, sad, surprise, and neutral. This system requires intensive research to address issues with human diversity, various unique human expressions, and variety of human facial features due to age differences. These issues generally affect the ability of the FER system to detect human emotions with high accuracy. Early stage of FER systems used simple supervised classification task algorithms like K-nearest neighbors (KNN) and artificial neural networks (ANN). These conventional FER systems have issues with low accuracy due to its inefficiency to extract significant features of several human emotions. To increase the accuracy of FER systems, deep learning (DL)-based methods, like convolutional neural networks (CNN), are proposed. These methods can find more complex features in the human face by means of the deeper connections within its architectures. However, the inference speed and computational costs of a DL-based FER system is often disregarded in exchange for higher accuracy results. To cope with this drawback, an optimized DL-based FER system is proposed in this study.An extreme version of Inception V3, known as Xception model, is leveraged by applying different network optimization methods. Specifically, network pruning and quantization are used to enable lower computational costs and reduce memory usage, respectively. To support low resource requirements, a 68-landmark face detector from Dlib is used in the early step of the FER system.Furthermore, a DL compiler is utilized to incorporate advanced optimization techniques to the Xception model to improve the inference speed of the FER system. In comparison to VGG-Net and ResNet50, the proposed optimized DL-based FER system experimentally demonstrates the objectives of the network optimization methods used. As a result, the proposed approach can be used to create an efficient and real-time FER system.

Keywords: deep learning, face detection, facial emotion recognition, network optimization methods

Procedia PDF Downloads 97

Authors: Jaime E. Fernandez, Pablo J. Valverde

Abstract:

This paper presents a methodological approach that aims to contrast/validate theoretical results from a corruption network game through probabilistic analysis of simulated microdata using Bayesian Networks (BNs). The research develops a public corruption model in a game theory framework. Theoretical results suggest a series of 'optimal settings' of model's exogenous parameters that boost the emergence of corruption. The paper contrasts these outcomes with probabilistic inference results based on BNs adjusted over simulated microdata. Principal findings indicate that probabilistic reasoning based on BNs significantly improves parameter specification and causal analysis in a public corruption game.

Keywords: Bayesian networks, probabilistic reasoning, public corruption, theoretical games

Procedia PDF Downloads 185

Authors: Ameen Jameel Alawneh

Abstract:

A Mobile ad-hoc network (MANET) is a collection of wireless mobile hosts that dynamically form a temporary network without the aid of a system administrator. It has neither fixed infrastructure nor wireless ad hoc sessions. It inherently reaches several nodes with a single transmission, and each node functions as both a host and a router. The network maybe represented as a set of clusters each managed by clusterhead. The cluster size is not fixed and it depends on the movement of nodes. We proposed a clusterhead size algorithm (CHSize). This clustering algorithm can be used by several routing algorithms for ad hoc networks. An elected clusterhead is assigned for communication with all other clusters. Analysis and simulation of the algorithm has been implemented using GloMoSim networks simulator, MATLAB and MAPL11 proved that the proposed algorithm achieves the goals.

Keywords: simulation, MANET, Ad-hoc, cluster head size, linked cluster algorithm, loss and dropped packets

Procedia PDF Downloads 369

Authors: Semih Demir, Anil Celebi

Abstract:

Circuit switched networks widely used until the end of the 20^th century have been transformed into packages switched networks. Time Division Multiplexing over Internet Protocol (TDMoIP) is a system that enables Time Division Multiplexing (TDM) traffic to be carried over packet switched networks (PSN). In TDMoIP systems, devices that send TDM data to the PSN and receive it from the network must operate with the same clock frequency. In this study, it was aimed to implement clock synchronization process in Field Programmable Gate Array (FPGA) chips using time information attached to the packages received from PSN. The designed hardware is verified using the datasets obtained for the different carrier types and comparing the results with the software model. Field tests are also performed by using the real time TDMoIP system.

Keywords: clock recovery on TDMoIP, FPGA, MATLAB reference model, clock synchronization

Procedia PDF Downloads 249

Authors: Y. Kourd, D. Lefebvre, N. Guersi

Abstract:

This paper presents a new idea for fault detection and isolation (FDI) technique which is applied to industrial system. This technique is based on Neural Networks fault-free and Faulty behaviors Models (NNFM's). NNFM's are used for residual generation, while decision tree architecture is used for residual evaluation. The decision tree is realized with data collected from the NNFM’s outputs and is used to isolate detectable faults depending on computed threshold. Each part of the tree corresponds to specific residual. With the decision tree, it becomes possible to take the appropriate decision regarding the actual process behavior by evaluating few numbers of residuals. In comparison to usual systematic evaluation of all residuals, the proposed technique requires less computational effort and can be used for on line diagnosis. An application example is presented to illustrate and confirm the effectiveness and the accuracy of the proposed approach.

Keywords: neural networks, decision trees, diagnosis, behaviors

Procedia PDF Downloads 469

Authors: Riyadh Alsultani, Ali Majdi

Abstract:

It is necessary to develop a design methodology that includes the possibility of seismic events occurring in a region, the vulnerability of the civil hydraulic structure, and the effects of the occurrence hazard on society, environment, and economy in order to evaluate the flood and earthquake risks of coastal bridge networks. This paper presents a design approach for the assessment of the risk and sustainability of coastal bridge networks under time-variant flood-earthquake conditions. The social, environmental, and economic indicators of the network are used to measure its sustainability. These consist of anticipated loss, downtime, energy waste, and carbon dioxide emissions. The design process takes into account the possibility of happening in a set of flood and earthquake scenarios that represent the local seismic activity. Based on the performance of each bridge as determined by fragility assessments, network linkages are measured. The network's connections and bridges' damage statuses after an earthquake scenario determine the network's sustainability and danger. The sustainability measures' temporal volatility and the danger of structural degradation are both highlighted. The method is shown using a transportation network in Baghdad, Iraq.

Keywords: sustainability, Coastal bridge networks, flood-earthquake risk, structural design

Procedia PDF Downloads 64

Authors: Ons Slimene, Aroua Taamallah, Maha Khemaja

Abstract:

In recent years, human activity recognition (HAR) has been a key area of research due to its diverse applications. It has garnered increasing attention in the field of computer vision. HAR plays an important role in people’s daily lives as it has the ability to learn advanced knowledge about human activities from data. In HAR, activities are usually represented by exploiting different types of sensors, such as embedded sensors or visual sensors. However, these sensors have limitations, such as local obstacles, image-related obstacles, sensor unreliability, and consumer concerns. Recently, several deep learning-based approaches have been proposed for HAR and these approaches are classified into two categories based on the type of data used: vision-based approaches and sensor-based approaches. This research paper highlights the importance of multimodal data fusion from skeleton data obtained from videos and data generated by embedded sensors using deep neural networks for achieving HAR. We propose a deep multimodal fusion network based on a twostream architecture. These two streams use the Convolutional Neural Network combined with the Bidirectional LSTM (CNN BILSTM) to process skeleton data and data generated by embedded sensors and the fusion at the feature level is considered. The proposed model was evaluated on a public OPPORTUNITY++ dataset and produced a accuracy of 96.77%.

Keywords: human activity recognition, action recognition, sensors, vision, human-centric sensing, deep learning, context-awareness

Procedia PDF Downloads 75