Search results for: deep capsule network

Authors: Omojokun G. Aju, Adedayo O. Sule

Abstract:

ZigBee wireless sensor and control network is one of the most popularly deployed wireless technologies in recent years. This is because ZigBee is an open standard lightweight, low-cost, low-speed, low-power protocol that allows true operability between systems. It is built on existing IEEE 802.15.4 protocol and therefore combines the IEEE 802.15.4 features and newly added features to meet required functionalities thereby finding applications in wide variety of wireless networked systems. ZigBee‘s current focus is on embedded applications of general-purpose, inexpensive, self-organising networks which requires low to medium data rates, high number of nodes and very low power consumption such as home/industrial automation, embedded sensing, medical data collection, smart lighting, safety and security sensor networks, and monitoring systems. Although the ZigBee design specification includes security features to protect data communication confidentiality and integrity, however, when simplicity and low-cost are the goals, security is normally traded-off. A lot of researches have been carried out on ZigBee technology in which emphasis has mainly been placed on ZigBee network performance characteristics such as energy efficiency, throughput, robustness, packet delay and delivery ratio in different scenarios and applications. This paper investigate and analyse the data accuracy, network implementation difficulties and security challenges of ZigBee network applications in star-based and mesh-based topologies with emphases on its home monitoring application using the ZigBee ProBee ZE-10 development boards for the network setup. The paper also expose some factors that need to be considered when designing ZigBee network applications and suggest ways in which ZigBee network can be designed to provide more resilient to network attacks.

Keywords: home monitoring, IEEE 802.14.5, topology, wireless security, wireless sensor network (WSN), ZigBee

Procedia PDF Downloads 376

Authors: Ci Lin, Tet Yeap, Iluju Kiringa

Abstract:

This paper introduces the subspace rotation algorithm (SRA) to train the Restricted Hopfield Network (RHN) as an auto-associative memory. Subspace rotation algorithm is a gradient-free subspace tracking approach based on the singular value decomposition (SVD). In comparison with Backpropagation Through Time (BPTT) on training RHN, it is observed that SRA could always converge to the optimal solution and BPTT could not achieve the same performance when the model becomes complex, and the number of patterns is large. The AUTS case study showed that the RHN model trained by SRA could achieve a better structure of attraction basin with larger radius(in general) than the Hopfield Network(HNN) model trained by Hebbian learning rule. Through learning 10000 patterns from MNIST dataset with RHN models with different number of hidden nodes, it is observed that an several components could be adjusted to achieve a balance between recovery accuracy and noise resistance.

Keywords: hopfield neural network, restricted hopfield network, subspace rotation algorithm, hebbian learning rule

Procedia PDF Downloads 113

Authors: Zhizhong Yuan, Lili Wang, Chenya Zheng, Wuqi Yang

Abstract:

How to encourage and support peer-to-peer (P2P) network lending, and effectively monitor the risk of P2P network lending, has become the focus of the Chinese government departments, industrialists, experts and scholars in recent years. The reason is that this convenient online micro-credit service brings a series of credit risks and other issues. Avoiding the risks brought by the P2P network lending model, it can better play a benign role and help China's small and medium-sized private enterprises with vigorous development to solve the capital needs; otherwise, it will bring confusion to the normal financial order. As a form of financial services, P2P network lending has injected new blood into China's non-government finance in the past ten years, and has found a way out for idle funds and made up for the shortage of traditional financial services in China. However, it lacks feasible measures in credit evaluation and government supervision. This paper collects a large amount of data about P2P network lending of China. The data collection comes from the official media of the Chinese government, the public achievements of existing researchers and the analysis and collation of correlation data by the authors. The research content of this paper includes literature review; the current situation of China's P2P network lending development; the risk analysis of P2P network lending in China; the risk prevention strategy of P2P network lending in China. The focus of this paper is to try to find a specific program to strengthen supervision and avoid risks from the perspective of government regulators, operators of P2P network lending platform, investors and users of funds. These main measures include: China needs to develop self-discipline organization of P2P network lending industry and formulate self-discipline norms as soon as possible; establish a regular information disclosure system of P2P network lending platform; establish censorship of credit rating of borrowers; rectify the P2P network lending platform in compliance through the implementation of bank deposition. The results and solutions will benefit all the P2P network lending platforms, creditors, debtors, bankers, independent auditors and government agencies of China and other countries.

Keywords: peer-to-peer(P2P), regulation, risk prevention, supervision

Procedia PDF Downloads 164

Authors: Katsumi Hirata

Abstract:

Certain systems can function well only if they recognize the sound environment as humans do. In this research, we focus on sound classification by adopting a convolutional neural network and aim to develop a method that automatically classifies various environmental sounds. Although the neural network is a powerful technique, the performance depends on the type of input data. Therefore, we propose an approach via a slice bispectrogram, which is a third-order spectrogram and is a slice version of the amplitude for the short-time bispectrum. This paper explains the slice bispectrogram and discusses the effectiveness of the derived method by evaluating the experimental results using the ESC‑50 sound dataset. As a result, the proposed scheme gives high accuracy and stability. Furthermore, some relationship between the accuracy and non-Gaussianity of sound signals was confirmed.

Keywords: environmental sound, bispectrum, spectrogram, slice bispectrogram, convolutional neural network

Procedia PDF Downloads 122

Authors: Tahani Aljohani, Jialin Yu, Alexandra. I. Cristea

Abstract:

The more an educational system knows about a learner, the more personalised interaction it can provide, which leads to better learning. However, asking a learner directly is potentially disruptive, and often ignored by learners. Especially in the booming realm of MOOC Massive Online Learning platforms, only a very low percentage of users disclose demographic information about themselves. Thus, in this paper, we aim to predict learners’ demographic characteristics, by proposing an approach using linguistically motivated Deep Learning Architectures for Learner Profiling, particularly targeting gender prediction on a FutureLearn MOOC platform. Additionally, we tackle here the difficult problem of predicting the gender of learners based on their comments only – which are often available across MOOCs. The most common current approaches to text classification use the Long Short-Term Memory (LSTM) model, considering sentences as sequences. However, human language also has structures. In this research, rather than considering sentences as plain sequences, we hypothesise that higher semantic - and syntactic level sentence processing based on linguistics will render a richer representation. We thus evaluate, the traditional LSTM versus other bleeding edge models, which take into account syntactic structure, such as tree-structured LSTM, Stack-augmented Parser-Interpreter Neural Network (SPINN) and the Structure-Aware Tag Augmented model (SATA). Additionally, we explore using different word-level encoding functions. We have implemented these methods on Our MOOC dataset, which is the most performant one comparing with a public dataset on sentiment analysis that is further used as a cross-examining for the models' results.

Keywords: deep learning, data mining, gender predication, MOOCs

Procedia PDF Downloads 145

Authors: Krisana Chueachainat

Abstract:

The study is about the identity, typology and symbol using in communication through offline and online social network of each Thai football Premier League supporters. Also, study is about the factors that affected the sport to become the growing business and the communication factors that play the important role in the growth of the sport business. The Thai Premier League communicated with supporters in order to show the identity of each supporter and club in different ways. The expression of the identity was shown through online social network and offline told other people who they were. The study also about the factor that impacted the roles and communication factors that make football become the growing business. The factor that impact to the growth of football to the business, if clubs can action, the sport business would be to higher level and also push Thailand’s football to be effective and equal to other countries.

Keywords: online social network, offline social network, Thai football, supporters

Procedia PDF Downloads 297

Authors: Tuğçe Yücel

Abstract:

A Wireless Sensor Network(WSN) is a collection of sensor nodes which are deployed randomly in an area for surveillance. Efficient utilization of limited battery energy of sensors for increased network lifetime as well as data security are major design objectives for WSN. Moreover secure transmission of data sensed to a base station for further processing. Producing multiple copies of data packets and sending them on different paths is one of the strategies for this purpose, which leads to redundant energy consumption and hence reduced network lifetime. In this work we develop a restricted multi-copy multipath strategy where data move through ‘frequently’ or ‘heavily’ used sensors is copied by the sensor incident to such central nodes and sent on node-disjoint paths. We develop a mixed integer programing(MIP) model and heuristic approach present some preleminary test results.

Keywords: MIP, sensor, telecommunications, WSN

Procedia PDF Downloads 506

Authors: Balogun R. Ayodeji, Adefisan E. Adesanya, Adeyewa Z. Debo, E. C. Okogbue

Abstract:

Deep convective systems during active periods of the West African monsoon season have not been properly investigated over better temporal and spatial resolution in West Africa. Deep convective systems are investigated over seven climatic zones of the West African sub-region, which are; west-coast rainforest, dry rainforest, Nigeria-Cameroon rainforest, Nigeria savannah, Central African and South Sudan (CASS) Savannah, Sudano-Sahel, and Sahel, using data from Tropical Rainfall Measurement Mission (TRMM) Precipitation Feature (PF) database. The vertical structure of the convective systems indicated by the presence of at least one 40 dBZ and reaching (attaining) at least 1km in the atmosphere showed strong core (highest frequency (%)) of reflectivity values around 2 km which is below the freezing level (4-5km) for all the zones. Echoes are detected above the 15km altitude much more frequently in the rainforest and Savannah zones than the Sudano and Sahel zones during active periods in March-May (MAM), whereas during active periods in June-September (JJAS) the savannahs, Sudano and Sahel zones convections tend to reach higher altitude more frequently than the rainforest zones. The percentage frequencies of deep convection indicated that the occurrences of the systems are within the range of 2.3-2.8% during both March-May (MAM) and June-September (JJAS) active periods in the rainforest and savannah zones. On the contrary, the percentage frequencies were found to be less than 2% in the Sudano and Sahel zones, except during the active-JJAS period in the Sudano zone.

Keywords: active periods, convective system, frequency, reflectivity

Procedia PDF Downloads 146

Authors: Amira E. Derbalah, Doaa M. Zaghloul

Abstract:

10 clinically healthy hemal nodes were collected from male bulls aged 2-3 years. Light microscopy revealed a capsule of connective tissue consisted mainly of collagen fiber surrounding hemal node, numerous erythrocytes were found in wide subcapsular sinus under the capsule. The parenchyma of the hemal node was divided into cortex and medulla. Diffused lymphocytes, and lymphoid follicles, having germinal centers were the main components of the cortex, while in the medulla there was wide medullary sinus, diffused lymphocytes and few lymphoid nodules. The area occupied with lymph nodules was larger than that occupied with non-nodular structure of lymphoid cords and blood sinusoids. Electron microscopy revealed the cellular components of hemal node including elements of circulating erythrocytes intermingled with lymphocytes, plasma cells, mast cells, reticular cells, macrophages, megakaryocytes and endothelial cells lining the blood sinuses. The lymphocytes were somewhat triangular in shape with cytoplasmic processes extending between adjacent erythrocytes. Nuclei were triangular to oval in shape, lightly stained with clear nuclear membrane indentation and clear nucleoli. The reticular cells were elongated in shape with cytoplasmic processes extending between adjacent lymphocytes, rough endoplasmic reticulum, ribosomes and few lysosomes were seen in their cytoplasm. Nucleus was elongated in shape with less condensed chromatin. Plasma cells were oval to irregular in shape with numerous dilated rough endoplasmic reticulum containing electron lucent material occupying the whole cytoplasm and few mitochondria were found. Nuclei were centrally located and oval in shape with heterochromatin emarginated and often clumped near the nuclear membrane. Occasionally megakaryocytes and mast cells were seen among lymphocytes. Megakaryocytes had multilobulated nucleus and free ribosomes often appearing as small aggregates in their cytoplasm, while mast cell had their characteristic electron dense granule in the cytoplasm, few electron lucent granules were found also, we conclude that, the main function of the hemal node of cattle is proliferation of lymphocytes. No role for plasma cell in erythrophagocytosis could be suggested.

Keywords: cattle, electron microscopy, hemal node, histology, immune system

Procedia PDF Downloads 397

Authors: Guangzhuang Yu, Chun-Chu Liu

Abstract:

With the advent of the Internet era in recent years, social music education has gradually shifted from the original entity education mode to the mode of entity plus network teaching. No matter for school music education, professional music education or social music education, the teaching quality is the most important evaluation index. Regarding the research on teaching quality evaluation, scholars at home and abroad have contributed a lot of research results on the basis of multiple methods and evaluation subjects. However, to our best knowledge the complete evaluation model for the virtual teaching interaction mode of the emerging network music education Application (APP) has not been established. This research firstly found out the basic dimensions that accord with the teaching quality required by the three parties, constructing the quality evaluation index system; and then, on the basis of expounding the connotation of each index, it determined the weight of each index by using method of fuzzy analytic hierarchy process, providing ideas and methods for scientific, objective and comprehensive evaluation of the teaching quality of network education APP.

Keywords: network music education APP, teaching quality evaluation, index and connotation

Procedia PDF Downloads 121

Authors: Nguyen J. M., Lucas G., Brunner M., Ruan S., Antonioli D.

Abstract:

Deep learning based on convolutional neural networks (CNN) is a very powerful technique for classifying information from an image. We propose a new method, DeepNic, to transform each variable of a tabular dataset into an image where each pixel represents a set of conditions that allow the variable to make an error-free prediction. The contrast of each pixel is proportional to its prediction performance and the color of each pixel corresponds to a sub-family of NICs. NICs are probabilities that depend on the number of inputs to each neuron and the range of coefficients of the inputs. Each variable can therefore be expressed as a function of a matrix of 2 vectors corresponding to an image whose pixels express predictive capabilities. Our objective is to transform each variable of tabular data into images into an image that can be analysed by CNNs, unlike other methods which use all the variables to construct an image. We analyse the NIC information of each variable and express it as a function of the number of neurons and the range of coefficients used. The predictive value and the category of the NIC are expressed by the contrast and the color of the pixel. We have developed a pipeline to implement this technology and have successfully applied it to genomic expressions on an Affymetrix chip.

Keywords: tabular data, deep learning, perfect trees, NICS

Procedia PDF Downloads 86

Authors: Edward Guillén, Jhordany Rodriguez, Rafael Páez

Abstract:

Anomaly detection techniques have been focused on two main components: data extraction and selection and the second one is the analysis performed over the obtained data. The goal of this paper is to analyze the influence that each of these components has over the system performance by evaluating detection over network scenarios with different setups. The independent variables are as follows: the number of system inputs, the way the inputs are codified and the complexity of the analysis techniques. For the analysis, some approaches of artificial neural networks are implemented with different number of layers. The obtained results show the influence that each of these variables has in the system performance.

Keywords: network intrusion detection, machine learning, artificial neural network, anomaly detection module

Procedia PDF Downloads 338

Authors: P. Prasanna Murali Krishna, M. V. Subramanyam, K. Satya Prasad

Abstract:

File sharing in networks are generally achieved using Peer-to-Peer (P2P) applications. Structured P2P approaches are widely used in adhoc networks due to its distributed and scalability features. Efficient mechanisms are required to handle the huge amount of data distributed to all peers. The intrinsic characteristics of P2P system makes for easier content distribution when compared to client-server architecture. All the nodes in a P2P network act as both client and server, thus, distributing data takes lesser time when compared to the client-server method. CHORD protocol is a resource routing based where nodes and data items are structured into a 1- dimensional ring. The structured lookup algorithm of Chord is advantageous for distributed P2P networking applications. Though, structured approach improves lookup performance in a high bandwidth wired network it could contribute to unnecessary overhead in overlay networks leading to degradation of network performance. In this paper, the performance of existing CHORD protocol on Wireless Mesh Network (WMN) when nodes are static and dynamic is investigated.

Keywords: wireless mesh network (WMN), structured P2P networks, peer to peer resource sharing, CHORD Protocol, DHT

Procedia PDF Downloads 478

Authors: J. K. Adedeji, S. T. Ijatuyi

Abstract:

The application of neural network using pattern recognition to study the fluid dynamics and predict the groundwater reservoirs properties has been used in this research. The essential of geophysical survey using the manual methods has failed in basement environment, hence the need for an intelligent computing such as predicted from neural network is inevitable. A non-linear neural network with an XOR (exclusive OR) output of 8-bits configuration has been used in this research to predict the nature of groundwater reservoirs and fluid dynamics of a typical basement crystalline rock. The control variables are the apparent resistivity of weathered layer (p₁), fractured layer (p₂), and the depth (h), while the dependent variable is the flow parameter (F=λ). The algorithm that was used in training the neural network is the back-propagation coded in C++ language with 300 epoch runs. The neural network was very intelligent to map out the flow channels and detect how they behave to form viable storage within the strata. The neural network model showed that an important variable g_r (gravitational resistance) can be deduced from the elevation and apparent resistivity p_a. The model results from SPSS showed that the coefficients, a, b and c are statistically significant with reduced standard error at 5%.

Keywords: gravitational resistance, neural network, non-linear, pattern recognition

Procedia PDF Downloads 209

Authors: Sheng Fu, Yinbo Gao, Hao Lin

Abstract:

In the current mine air compressors monitoring system, there are some difficulties in the installation and maintenance because of the wired connection. To solve the problem, this paper introduces a new air compressors monitoring system based on ZigBee in which the monitoring parameters are transmitted wirelessly. The collecting devices are designed to form a cluster network to collect vibration, temperature, and pressure of air cylinders and other parameters. All these devices are battery-powered. Besides, the monitoring software in PC is developed using MFC. Experiments show that the designed wireless sensor network works well in the site environmental condition and the system is very convenient to be installed since the wireless connection. This monitoring system will have a wide application prospect in the upgrade of the old monitoring system of the air compressors.

Keywords: condition monitoring, wireless sensor network, air compressor, zigbee, data collecting

Procedia PDF Downloads 501

Authors: Venkata Narayana Chejarla, Nafisa Parvez Shaik, Gopi Vara Prasad Marabathula, Deva Kumar Bejjam

Abstract:

Yoga is an advanced, well-recognized method with roots in Indian philosophy. Yoga benefits both the body and the psyche. Yoga is a regular exercise that helps people relax and sleep better while also enhancing their balance, endurance, and concentration. Yoga can be learned in a variety of settings, including at home with the aid of books and the internet as well as in yoga studios with the guidance of an instructor. Self-learning does not teach the proper yoga poses, and doing them without the right instruction could result in significant injuries. We developed "Online Yoga Asana Trainer using Deep Learning" so that people could practice yoga without a teacher. Our project is developed using Tensorflow, Movenet, and Keras models. The system makes use of data from Kaggle that includes 25 different yoga poses. The first part of the process involves applying the movement model for extracting the 17 key points of the body from the dataset, and the next part involves preprocessing, which includes building a pose classification model using neural networks. The system scores a 98.3% accuracy rate. The system is developed to work with live videos.

Keywords: yoga, deep learning, movenet, tensorflow, keras, CNN

Procedia PDF Downloads 238

Authors: Yanjun Chen, Chuanping Hu, Jie Shao, Lin Mei, Chongyang Zhang

Abstract:

Traditional image classification is based on encoding scheme (e.g. Fisher Vector, Vector of Locally Aggregated Descriptor) with low-level image features (e.g. SIFT, HoG). Compared to these low-level local features, deep convolutional features obtained at the mid-level layer of convolutional neural networks (CNN) have richer information but lack of geometric invariance. For scene classification, there are scattered objects with different size, category, layout, number and so on. It is crucial to find the distinctive objects in scene as well as their co-occurrence relationship. In this paper, we propose a method to take advantage of both deep convolutional features and the traditional encoding scheme while taking object-centric and scene-centric information into consideration. First, to exploit the object-centric and scene-centric information, two CNNs that trained on ImageNet and Places dataset separately are used as the pre-trained models to extract deep convolutional features at multiple scales. This produces dense local activations. By analyzing the performance of different CNNs at multiple scales, it is found that each CNN works better in different scale ranges. A scale-wise CNN adaption is reasonable since objects in scene are at its own specific scale. Second, a fisher kernel is applied to aggregate a global representation at each scale and then to merge into a single vector by using a post-processing method called scale-wise normalization. The essence of Fisher Vector lies on the accumulation of the first and second order differences. Hence, the scale-wise normalization followed by average pooling would balance the influence of each scale since different amount of features are extracted. Third, the Fisher vector representation based on the deep convolutional features is followed by a linear Supported Vector Machine, which is a simple yet efficient way to classify the scene categories. Experimental results show that the scale-specific feature extraction and normalization with CNNs trained on object-centric and scene-centric datasets can boost the results from 74.03% up to 79.43% on MIT Indoor67 when only two scales are used (compared to results at single scale). The result is comparable to state-of-art performance which proves that the representation can be applied to other visual recognition tasks.

Keywords: deep convolutional features, Fisher Vector, multiple scales, scale-specific normalization

Procedia PDF Downloads 327

Authors: Ali Mohammad Mobaser Azad, Alireza Akhlaghinia

Abstract:

Blockchain is changing our world and launching a new generation of decentralized networks. Meanwhile, Blockchain-based networks like Ethereum have been created and they will facilitate these processes using tools like smart contracts. The Ethereum has fundamental structures, each of which affects the activity of the nodes. Our purpose in this paper is to review similar research and examine various components to demonstrate the performance of the Ethereum network and to do this, and we used the data published by the Ethereum Foundation in different time spots to examine the number of changes that determine the status of network performance. This will help other researchers understand better Ethereum in different situations.

Keywords: blockchain, ethereum, smart contract, decentralization consensus algorithm

Procedia PDF Downloads 220

Authors: Rania Alshikhe, Vinita Jindal

Abstract:

Digital innovation has played a crucial role in managing smart transportation. For this, big trajectory data collected from traveling vehicles, such as taxis through installed global positioning system (GPS)-enabled devices can be utilized. It offers an unprecedented opportunity to trace the movements of vehicles in fine spatiotemporal granularity. This paper aims to explore big trajectory data to measure the travel efficiency of road networks using the proposed statistical travel efficiency measure (STEM) across an entire city. Further, it identifies the cause of low travel efficiency by proposed least square approximation network-based causality exploration (LANCE). Finally, the resulting data analysis reveals the causes of low travel efficiency, along with the road segments that need to be optimized to improve the traffic conditions and thus minimize the average travel time from given point A to point B in the road network. Obtained results show that our proposed approach outperforms the baseline algorithms for measuring the travel efficiency of the road network.

Keywords: GPS trajectory, road network, taxi trips, digital map, big data, STEM, LANCE

Procedia PDF Downloads 154

Authors: Joseph Gear, Yue Xu, Ernest Foo, Praveen Gauravaran, Zahra Jadidi, Leonie Simpson

Abstract:

Deep learning methods have been seeing an increasing application to the long-standing security research goal of automatic vulnerability detection for source code. Attention, however, must still be paid to the task of producing vector representations for source code (code embeddings) as input for these deep learning models. Graphical representations of code, most predominantly Abstract Syntax Trees and Code Property Graphs, have received some use in this task of late; however, for very large graphs representing very large code snip- pets, learning becomes prohibitively computationally expensive. This expense may be reduced by intelligently pruning this input to only vulnerability-relevant information; however, little research in this area has been performed. Additionally, most existing work comprehends code based solely on the structure of the graph at the expense of the information contained by the node in the graph. This paper proposes Semantic-enhanced Code Embedding for Vulnerability Discovery (SCEVD), a deep learning model which uses semantic-based feature selection for its vulnerability classification model. It uses information from the nodes as well as the structure of the code graph in order to select features which are most indicative of the presence or absence of vulnerabilities. This model is implemented and experimentally tested using the SARD Juliet vulnerability test suite to determine its efficacy. It is able to improve on existing code graph feature selection methods, as demonstrated by its improved ability to discover vulnerabilities.

Keywords: code representation, deep learning, source code semantics, vulnerability discovery

Procedia PDF Downloads 153

Authors: Adel Bougamouza, Boualam Remini, Abd El Hadi Ammari, Feteh Sakhraoui

Abstract:

The assessment of sediments being carried by a river is importance for planning and designing of various water resources projects. In this study, Artificial Neural Network Techniques are used to estimate the daily suspended sediment concentration for the corresponding daily discharge flow in the upstream of Foum El Gherza dam, Biskra, Algeria. The FFNN, GRNN, and RBNN models are established for estimating current suspended sediment values. Some statistics involving RMSE and R2 were used to evaluate the performance of applied models. The comparison of three AI models showed that the RBNN model performed better than the FFNN and GRNN models with R2 = 0.967 and RMSE= 5.313 mg/l. Therefore, the ANN model had capability to improve nonlinear relationships between discharge flow and suspended sediment with reasonable precision.

Keywords: artificial neural network, Oued Abiod watershed, feedforward network, generalized regression network, radial basis network, sediment concentration

Procedia PDF Downloads 413

Authors: Armin Nabaei, M. Omair Ahmad, M. N. S. Swamy

Abstract:

Many facial expression and emotion recognition methods in the traditional approaches of using LDA, PCA, and EBGM have been proposed. In recent years deep learning models have provided a unique platform addressing by automatically extracting the features for the detection of facial expression and emotions. However, deep networks require large training datasets to extract automatic features effectively. In this work, we propose an efficient emotion detection algorithm using face images when only small datasets are available for training. We design a deep network whose feature extraction capability is enhanced by utilizing several parallel modules between the input and output of the network, each focusing on the extraction of different types of coarse features with fined grained details to break the symmetry of produced information. In fact, we leverage long range dependencies, which is one of the main drawback of CNNs. We develop this work by introducing a Dynamic Soft-Margin SoftMax.The conventional SoftMax suffers from reaching to gold labels very soon, which take the model to over-fitting. Because it’s not able to determine adequately discriminant feature vectors for some variant class labels. We reduced the risk of over-fitting by using a dynamic shape of input tensor instead of static in SoftMax layer with specifying a desired Soft- Margin. In fact, it acts as a controller to how hard the model should work to push dissimilar embedding vectors apart. For the proposed Categorical Loss, by the objective of compacting the same class labels and separating different class labels in the normalized log domain.We select penalty for those predictions with high divergence from ground-truth labels.So, we shorten correct feature vectors and enlarge false prediction tensors, it means we assign more weights for those classes with conjunction to each other (namely, “hard labels to learn”). By doing this work, we constrain the model to generate more discriminate feature vectors for variant class labels. Finally, for the proposed optimizer, our focus is on solving weak convergence of Adam optimizer for a non-convex problem. Our noteworthy optimizer is working by an alternative updating gradient procedure with an exponential weighted moving average function for faster convergence and exploiting a weight decay method to help drastically reducing the learning rate near optima to reach the dominant local minimum. We demonstrate the superiority of our proposed work by surpassing the first rank of three widely used Facial Expression Recognition datasets with 93.30% on FER-2013, and 16% improvement compare to the first rank after 10 years, reaching to 90.73% on RAF-DB, and 100% k-fold average accuracy for CK+ dataset, and shown to provide a top performance to that provided by other networks, which require much larger training datasets.

Keywords: computer vision, facial expression recognition, machine learning, algorithms, depp learning, neural networks

Procedia PDF Downloads 73

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 274

Authors: Vishal H. Shukla, Sanjay B. Deshmukh

Abstract:

Traffic engineering, at its center, is the ability of moving traffic approximately so that traffic from a congested link is moved onto the unused capacity on another link. Traffic Engineering ensures the best possible use of the resources. Now to support traffic engineering in the today’s network, Multiprotocol Label Switching (MPLS) is being used which is very helpful for reliable packets delivery in an ongoing internet services. Here a topology is been implemented on GNS3 to focus on the analysis of the communication take place from one site to other through the ISP. The comparison is made between the IP network & MPLS network based on Bandwidth & Jitter which are one of the performance parameters using JPERF simulator.

Keywords: GNS3, JPERF, MPLS, traffic engineering, VMware

Procedia PDF Downloads 481

Authors: Afaneen Anwer, Samara M. Kamil

Abstract:

Load Forecast required prediction accuracy based on optimal operation and maintenance. A good accuracy is the basis of economic dispatch, unit commitment, and system reliability. A good load forecasting system fulfilled fast speed, automatic bad data detection, and ability to access the system automatically to get the needed data. In this paper, the formulation of the load forecasting is discussed and the solution is obtained by using artificial neural network method. A MATLAB environment has been used to solve the load forecasting schedule of Iraqi super grid network considering the daily load for three years. The obtained results showed a good accuracy in predicting the forecasted load.

Keywords: load forecasting, neural network, back-propagation algorithm, Iraqi power system

Procedia PDF Downloads 574

Authors: Budhvin T. Withana, Sulochana Rupasinghe

Abstract:

The educational system faces a significant concern with regards to Dyslexia and Dysgraphia, which are learning disabilities impacting reading and writing abilities. This is particularly challenging for children who speak the Sinhala language due to its complexity and uniqueness. Commonly used methods to detect the risk of Dyslexia and Dysgraphia rely on subjective assessments, leading to limited coverage and time-consuming processes. Consequently, delays in diagnoses and missed opportunities for early intervention can occur. To address this issue, the project developed a hybrid model that incorporates various deep learning techniques to detect the risk of Dyslexia and Dysgraphia. Specifically, Resnet50, VGG16, and YOLOv8 models were integrated to identify handwriting issues. The outputs of these models were then combined with other input data and fed into an MLP model. Hyperparameters of the MLP model were fine-tuned using Grid Search CV, enabling the identification of optimal values for the model. This approach proved to be highly effective in accurately predicting the risk of Dyslexia and Dysgraphia, providing a valuable tool for early detection and intervention. The Resnet50 model exhibited a training accuracy of 0.9804 and a validation accuracy of 0.9653. The VGG16 model achieved a training accuracy of 0.9991 and a validation accuracy of 0.9891. The MLP model demonstrated impressive results with a training accuracy of 0.99918, a testing accuracy of 0.99223, and a loss of 0.01371. These outcomes showcase the high accuracy achieved by the proposed hybrid model in predicting the risk of Dyslexia and Dysgraphia.

Keywords: neural networks, risk detection system, dyslexia, dysgraphia, deep learning, learning disabilities, data science

Procedia PDF Downloads 59

Authors: Boukari Nassim

Abstract:

This paper describes the use of odd pair autoregressive coefficients (Yule _Walker and Burg) for the feature extraction of electroencephalogram (EEG) signals. In the classification: the radial basis function neural network neural network (RBFNN) is employed. The RBFNN is described by his architecture and his characteristics: as the RBF is defined by the spread which is modified for improving the results of the classification. Five types of EEG signals are defined for this work: Set A, Set B for normal signals, Set C, Set D for interictal signals, set E for ictal signal (we can found that in Bonn university). In outputs, two classes are given (AC, AD, AE, BC, BD, BE, CE, DE), the best accuracy is calculated at 99% for the combined odd pair autoregressive coefficients. Our method is very effective for the diagnosis of epileptic EEG signals.

Keywords: epilepsy, EEG signals classification, combined odd pair autoregressive coefficients, radial basis function neural network

Procedia PDF Downloads 341

Authors: Derejaw Lake Melie, Alemu Kumlachew Tegegne

Abstract:

The increasing prevalence of social media in Ethiopia has exacerbated societal challenges by fostering the proliferation of negative emotional posts and comments. Illicit use of social media has further exacerbated divisions among the population. Addressing these issues through manual identification and aggregation of emotions from millions of users for swift decision-making poses significant challenges, particularly given the rapid growth of Amharic language usage on social platforms. Consequently, there is a critical need to develop an intelligent system capable of automatically detecting and categorizing negative emotional content into social, religious, and political categories while also filtering out toxic online content. This paper aims to leverage sentiment analysis techniques to achieve automatic detection and filtering of negative emotional content from Amharic social media texts, employing a comparative study of deep learning algorithms. The study utilized a dataset comprising 29,962 comments collected from social media platforms using comment exporter software. Data pre-processing techniques were applied to enhance data quality, followed by the implementation of deep learning methods for training, testing, and evaluation. The results showed that CNN, GRU, LSTM, and Bi-LSTM classification models achieved accuracies of 83%, 50%, 84%, and 86%, respectively. Among these models, Bi-LSTM demonstrated the highest accuracy of 86% in the experiment.

Keywords: negative emotion, emotion detection, social media filtering sentiment analysis, deep learning.

Procedia PDF Downloads 14

Authors: Zeta Dooly, Aidan Duane

Abstract:

The collaborative European funded research and development landscape provides prime environmental conditions for multi-disciplinary teams to learn and enhance their knowledge beyond the capability of training and learning within their own organisation cocoons. Whilst the emergence of the academic entrepreneur has changed the focus of educational institutions to that of quasi-businesses, the training and professional development of lecturers and academic staff are often not formalised to the same level as industry. This research focuses on industry and academic collaborative research funded by the European Commission. The impact of research is scalable if an optimum research network is created and managed effectively. This paper investigates network embeddedness, the nature of relationships, links, and nodes within a research network, and the enhancement of the network’s knowledge. The contribution of this paper extends our understanding of establishing and maintaining effective collaborative research networks. The effects of network embeddedness are recognized in the literature as pertinent to innovation and the economy. Network theory literature claims that networks are essential to innovative clusters such as Silicon valley and innovation in high tech industries. This research provides evidence to support the impact collaborative research has on the disparate individuals toward their innovative contributions to their organisations and their own professional development. This study adopts a qualitative approach and uncovers some of the challenges of multi-disciplinary research through case study insights. The contribution of this paper recommends the establishment of scaffolding to accommodate cooperation in research networks, role appointment, and addressing contextual complexities early to avoid problem cultivation. Furthermore, it suggests recommendations in relation to network formation, intra-network challenges in relation to open data, competition, friendships, and competency enhancement. The network capability is enhanced by the adoption of the relevant theories; network theory, open innovation, and social exchange, with the understanding that the network structure has an impact on innovation and social exchange in research networks. The research concludes that there is an opportunity to deepen our understanding of the impact of network reuse and network hoping that provides scaffolding for the network members to enhance and build upon their knowledge using a progressive approach.

Keywords: research networks, competency building, network theory, case study

Procedia PDF Downloads 122

Authors: Wanida Suwunniponth

Abstract:

The purpose of this research was to investigate the relationship between the factors of using online social network for tourism destination in case of Bangkok area in Thailand, by extending the use of technology acceptance model (TAM). This study employed by quantitative research and the target population were entrepreneurs and local people in Bangkok who use social network-Facebook concerning tourist destinations in Bangkok. Questionnaire was used to collect data from 300 purposive samples. The multiple regression analysis and path analysis were used to analyze data. The results revealed that most people who used Facebook for promoting tourism destinations in Bangkok perceived ease of use, perceived usefulness, perceived trust in using Facebook and influenced by social normative as well as having positive attitude towards using this application. Addition, the hypothesis results indicate that acceptance of online social network-Facebook was related to the positive attitude towards using of Facebook and related to their intention to use this application for tourism.

Keywords: Facebook, online social network, technology acceptance model, tourism destination

Procedia PDF Downloads 342