Search results for: graph attention neural network

Authors: Pius Marthin, Nihal Ata Tutkun

Abstract:

Survival analysis has become one of the paramount procedures in the modeling of time-to-event data. When we encounter complex survival problems, the traditional approach remains limited in accounting for the complex correlational structure between the covariates and the outcome due to the strong assumptions that limit the inference and prediction ability of the resulting models. Several studies exist on the deep learning approach to survival modeling; moreover, the application for the case of complex survival problems still needs to be improved. In addition, the existing models need to address the data structure's complexity fully and are subject to noise and redundant information. In this study, we design a deep learning technique (CmpXRnnSurv_AE) that obliterates the limitations imposed by traditional approaches and addresses the above issues to jointly predict the risk-specific probabilities and survival function for recurrent events with competing risks. We introduce the component termed Risks Information Weights (RIW) as an attention mechanism to compute the weighted cumulative incidence function (WCIF) and an external auto-encoder (ExternalAE) as a feature selector to extract complex characteristics among the set of covariates responsible for the cause-specific events. We train our model using synthetic and real data sets and employ the appropriate metrics for complex survival models for evaluation. As benchmarks, we selected both traditional and machine learning models and our model demonstrates better performance across all datasets.

Keywords: cumulative incidence function (CIF), risk information weight (RIW), autoencoders (AE), survival analysis, recurrent events with competing risks, recurrent neural networks (RNN), long short-term memory (LSTM), self-attention, multilayers perceptrons (MLPs)

Procedia PDF Downloads 90

Authors: Daniel Hyun

Abstract:

Dental implants are one of the top 3 reasons to sue a dentist for malpractice. It involves dental implant complications, usually because of the angle of the implant from the surgery. At present, surgeons usually use a 3D-printed navigator that is customized for the patient’s teeth. However, those can’t be reused for other patients as they require time. Therefore, I made a guiding device to assist the surgeon in implant operations. The surgeon can input the objective of the operation, and the device constantly checks if the surgery is heading towards the objective within the set range, telling the surgeon by manipulating the LED. We tested the prototypes’ consistency and accuracy by checking the graph, average standard deviation, and the average change of the calculated angles. The accuracy of performance was also acquired by running the device and checking the outputs. My first prototype used accelerometer and gyroscope sensors from the Arduino MPU6050 sensor, getting a changeable graph, achieving 0.0295 of standard deviations, 0.25 of average change, and 66.6% accuracy of performance. The second prototype used only the gyroscope, and it got a constant graph, achieved 0.0062 of standard deviation, 0.075 of average change, and 100% accuracy of performance, indicating that the accelerometer sensor aggravated the functionality of the device. Using the gyroscope sensor allowed it to measure the orientations of separate axes without affecting each other and also increased the stability and accuracy of the measurements.

Keywords: implant, guide, accelerometer, gyroscope, handpiece

Procedia PDF Downloads 43

Authors: Suraj Mehrotra

Abstract:

The stock market is a volatile, bustling marketplace that is a cornerstone of economics. It defines whether companies are successful or in spiral. A thorough understanding of it is important - many companies have whole divisions dedicated to analysis of both their stock and of rivaling companies. Linking the world of finance and artificial intelligence (AI), especially the stock market, has been a relatively recent development. Predicting how stocks will do considering all external factors and previous data has always been a human task. With the help of AI, however, machine learning models can help us make more complete predictions in financial trends. Taking a look at the stock market specifically, predicting the open, closing, high, and low prices for the next day is very hard to do. Machine learning makes this task a lot easier. A model that builds upon itself that takes in external factors as weights can predict trends far into the future. When used effectively, new doors can be opened up in the business and finance world, and companies can make better and more complete decisions. This paper explores the various techniques used in the prediction of stock prices, from traditional statistical methods to deep learning and neural networks based approaches, among other methods. It provides a detailed analysis of the techniques and also explores the challenges in predictive analysis. For the accuracy of the testing set, taking a look at four different models - linear regression, neural network, decision tree, and naïve Bayes - on the different stocks, Apple, Google, Tesla, Amazon, United Healthcare, Exxon Mobil, J.P. Morgan & Chase, and Johnson & Johnson, the naïve Bayes model and linear regression models worked best. For the testing set, the naïve Bayes model had the highest accuracy along with the linear regression model, followed by the neural network model and then the decision tree model. The training set had similar results except for the fact that the decision tree model was perfect with complete accuracy in its predictions, which makes sense. This means that the decision tree model likely overfitted the training set when used for the testing set.

Keywords: machine learning, testing set, artificial intelligence, stock analysis

Procedia PDF Downloads 95

Authors: Gulfam Haider, sana danish

Abstract:

Convolutional Neural Networks (CNNs) have emerged as powerful tools for image classification, and the choice of optimizers profoundly affects their performance. The study of optimizers and their adaptations remains a topic of significant importance in machine learning research. While numerous studies have explored and advocated for various optimizers, the efficacy of these optimization techniques is still subject to scrutiny. This work aims to address the challenges surrounding the effectiveness of optimizers by conducting a comprehensive analysis and evaluation. The primary focus of this investigation lies in examining the performance of different optimizers when employed in conjunction with the popular activation function, Rectified Linear Unit (ReLU). By incorporating ReLU, known for its favorable properties in prior research, the aim is to bolster the effectiveness of the optimizers under scrutiny. Specifically, we evaluate the adjustment of these optimizers with both the original Softmax activation function and the modified ReLU activation function, carefully assessing their impact on overall performance. To achieve this, a series of experiments are conducted using a well-established benchmark dataset for image classification tasks, namely the Canadian Institute for Advanced Research dataset (CIFAR-10). The selected optimizers for investigation encompass a range of prominent algorithms, including Adam, Root Mean Squared Propagation (RMSprop), Adaptive Learning Rate Method (Adadelta), Adaptive Gradient Algorithm (Adagrad), and Stochastic Gradient Descent (SGD). The performance analysis encompasses a comprehensive evaluation of the classification accuracy, convergence speed, and robustness of the CNN models trained with each optimizer. Through rigorous experimentation and meticulous assessment, we discern the strengths and weaknesses of the different optimization techniques, providing valuable insights into their suitability for image classification tasks. By conducting this in-depth study, we contribute to the existing body of knowledge surrounding optimizers in CNNs, shedding light on their performance characteristics for image classification. The findings gleaned from this research serve to guide researchers and practitioners in making informed decisions when selecting optimizers and activation functions, thus advancing the state-of-the-art in the field of image classification with convolutional neural networks.

Keywords: deep neural network, optimizers, RMsprop, ReLU, stochastic gradient descent

Procedia PDF Downloads 125

Authors: Ian O'Loughlin

Abstract:

Memory networks aim to integrate some of the recent successes in machine learning with a dynamic memory base that can be updated and deployed in artificial reasoning tasks. These models involve training networks to identify, update, and operate over stored elements in a large memory array in order, for example, to ably perform question and answer tasks parsing real-world and simulated discourses. This family of approaches still faces numerous challenges: the performance of these network models in simulated domains remains considerably better than in open, real-world domains, wide-context cues remain elusive in parsing words and sentences, and even moderately complex sentence structures remain problematic. This innovation, employing an array of stored and updatable ‘memory’ elements over which the system operates as it parses text input and develops responses to questions, is a compelling one for at least two reasons: first, it addresses one of the difficulties that standard machine learning techniques face, by providing a way to store a large bank of facts, offering a way forward for the kinds of long-term reasoning that, for example, recurrent neural networks trained on a corpus have difficulty performing. Second, the addition of a stored long-term memory component in artificial reasoning seems psychologically plausible; human reasoning appears replete with invocations of long-term memory, and the stored but dynamic elements in the arrays of memory networks are deeply reminiscent of the way that human memory is readily and often characterized. However, this apparent psychological plausibility is belied by a recent turn in the study of human memory in cognitive science. In recent years, the very notion that there is a stored element which enables remembering, however dynamic or reconstructive it may be, has come under deep suspicion. In the wake of constructive memory studies, amnesia and impairment studies, and studies of implicit memory—as well as following considerations from the cognitive neuroscience of memory and conceptual analyses from the philosophy of mind and cognitive science—researchers are now rejecting storage and retrieval, even in principle, and instead seeking and developing models of human memory wherein plasticity and dynamics are the rule rather than the exception. In these models, storage is entirely avoided by modeling memory using a recurrent neural network designed to fit a preconceived energy function that attains zero values only for desired memory patterns, so that these patterns are the sole stable equilibrium points in the attractor network. So although the array of long-term memory elements in memory networks seem psychologically appropriate for reasoning systems, they may actually be incurring difficulties that are theoretically analogous to those that older, storage-based models of human memory have demonstrated. The kind of emergent stability found in the attractor network models more closely fits our best understanding of human long-term memory than do the memory network arrays, despite appearances to the contrary.

Keywords: artificial reasoning, human memory, machine learning, neural networks

Procedia PDF Downloads 271

Authors: Nasrin Mohammadhasani, Rosa Angela Fabio

Abstract:

The attention to relevance information is the key element for learning. Otherwise, Attention Deficit Hyperactivity Disorder (ADHD) students have a fuzzy visual pattern that prevents them to attention and remember learning subject. The present study aimed to test the hypothesis that the presence of a pedagogical agent can effectively support ADHD learner's attention and learning outcomes in a multimedia learning environment. The learning environment was integrated with a pedagogical agent, named Koosha as a social peer. This study employed a pretest and posttest experimental design with control group. The statistical population was 30 boys students, age 10-11 with ADHD that randomly assigned to learn with/without an agent in well designed environment for mathematic. The results suggested that experimental and control groups show a significant difference in time when they participated and mathematics achievement. According to this research, using the pedagogical agent can enhance learning of ADHD students by gaining and guiding their attention to relevance information part on display, so it can be considered as asocial cue that provides theme cognitive supports.

Keywords: attention, computer assisted instruction, multimedia learning environment, pedagogical agent

Procedia PDF Downloads 314

Authors: Liang Bai, Yandong Xiao, Haorang Wang, Songyang Lao

Abstract:

The directedness of links is crucial to determine the controllability in complex networks. Even the edge directions can determine the controllability of complex networks. Obviously, for a given network, we wish to design its edge directions that make this network approach the optimal controllability. In this work, we firstly introduce two methods to enhance network by assigning edge directions. However, these two methods could not completely mitigate the negative effects of inaccessibility and dilations. Thus, to approach the optimal network controllability, the edge directions must mitigate the negative effects of inaccessibility and dilations as much as possible. Finally, we propose the edge direction for optimal controllability. The optimal method has been found to be successfully useful on real-world and synthetic networks.

Keywords: complex network, dynamics, network control, optimization

Procedia PDF Downloads 185

Authors: Alper Taner, Yesim Benal Oztekin, Huseyin Duran

Abstract:

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

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

Procedia PDF Downloads 178

Authors: Devadrita Dey Sarkar

Abstract:

Regardless of the many technologic advances in the past decade, increased training and experience, and the obvious benefits of uniform standards, the false-negative rate in screening mammography remains unacceptably high .A computer aided neural network classification of regions of suspicion (ROS) on digitized mammograms is presented in this abstract which employs features extracted by a new technique based on independent component analysis. CAD is a concept established by taking into account equally the roles of physicians and computers, whereas automated computer diagnosis is a concept based on computer algorithms only. With CAD, the performance by computers does not have to be comparable to or better than that by physicians, but needs to be complementary to that by physicians. In fact, a large number of CAD systems have been employed for assisting physicians in the early detection of breast cancers on mammograms. A CAD scheme that makes use of lateral breast images has the potential to improve the overall performance in the detection of breast lumps. Because breast lumps can be detected reliably by computer on lateral breast mammographs, radiologists’ accuracy in the detection of breast lumps would be improved by the use of CAD, and thus early diagnosis of breast cancer would become possible. In the future, many CAD schemes could be assembled as packages and implemented as a part of PACS. For example, the package for breast CAD may include the computerized detection of breast nodules, as well as the computerized classification of benign and malignant nodules. In order to assist in the differential diagnosis, it would be possible to search for and retrieve images (or lesions) with these CAD systems, which would be reliable and useful method for quantifying the similarity of a pair of images for visual comparison by radiologists.

Keywords: CAD(computer-aided design), lesions, neural network, ROS(region of suspicion)

Procedia PDF Downloads 456

Authors: Maria Pavlova

Abstract:

In our days, it is possible to put the camera on different vehicles like quadcopter, train, airplane and etc. The camera also can be the input sensor in many different systems. That means the object recognition like non separate part of monitoring control can be key part of the most intelligent systems. The aim of this paper is to focus of the object recognition process during vehicles movement. During the vehicle’s movement the camera takes pictures from the environment without storage in Data Base. In case the camera detects a special object (for example human or animal), the system saves the picture and sends it to the work station in real time. This functionality will be very useful in emergency or security situations where is necessary to find a specific object. In another application, the camera can be mounted on crossroad where do not have many people and if one or more persons come on the road, the traffic lights became the green and they can cross the road. In this papers is presented the system has solved the aforementioned problems. It is presented architecture of the object recognition system includes the camera, Raspberry platform, GPS system, neural network, software and Data Base. The camera in the system takes the pictures. The object recognition is done in real time using the OpenCV library and Raspberry microcontroller. An additional feature of this library is the ability to display the GPS coordinates of the captured objects position. The results from this processes will be sent to remote station. So, in this case, we can know the location of the specific object. By neural network, we can learn the module to solve the problems using incoming data and to be part in bigger intelligent system. The present paper focuses on the design and integration of the image recognition like a part of smart systems.

Keywords: camera, object recognition, OpenCV, Raspberry

Procedia PDF Downloads 218

Authors: Zeina Bou Diab, Arij Daou

Abstract:

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

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

Procedia PDF Downloads 70

Authors: Hongyun Li, Zhibin Jiang

Abstract:

The travel of passengers in the urban rail transit network is influenced by changes in network structure and operational status, and the response of individual travel preferences to these changes also varies. Firstly, the influence of the suspension of urban rail transit line sections on passenger travel along the line is analyzed. Secondly, passenger travel trajectories containing multi-dimensional semantics are described based on network UD data. Next, passenger panel data based on spatio-temporal sequences is constructed to achieve frequent passenger clustering. Then, the Graph Convolutional Network (GCN) is used to model and identify the changes in travel modes of different types of frequent passengers. Finally, taking Shanghai Metro Line 11 as an example, the travel behavior patterns of frequent passengers after the Huaqiao section shutdown during the COVID-19 epidemic are analyzed. The results showed that after the section shutdown, most passengers would transfer to the nearest Anting station for boarding, while some passengers would transfer to other stations for boarding or cancel their travels directly. Among the passengers who transferred to Anting station for boarding, most of passengers maintained the original normalized travel mode, a small number of passengers waited for a few days before transferring to Anting station for boarding, and only a few number of passengers stopped traveling at Anting station or transferred to other stations after a few days of boarding on Anting station. The results can provide a basis for understanding urban rail transit passenger travel patterns and improving the accuracy of passenger flow prediction in abnormal operation scenarios.

Keywords: urban rail transit, section shutdown, frequent passenger, travel behavior pattern

Procedia PDF Downloads 84

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 392

Authors: John Onyima, Ikechukwu Ezepue

Abstract:

Virtualization and cloud computing are among the fast-growing computing innovations in recent times. Organisations all over the world are moving their computing services towards the cloud this is because of its rapid transformation of the organization’s infrastructure and improvement of efficient resource utilization and cost reduction. However, this technology brings new security threats and challenges about safety, reliability and data confidentiality. Evidently, no single security technique can guarantee security or protection against malicious attacks on a cloud computing network hence an integrated model of intrusion detection and prevention system has been proposed. Anomaly-based and signature-based detection techniques will be integrated to enable the network and its host defend themselves with some level of intelligence. The anomaly-base detection was implemented using the local deviation factor graph-based (LDFGB) algorithm while the signature-based detection was implemented using the snort algorithm. Results from this collaborative intrusion detection and prevention techniques show robust and efficient security architecture for cloud computing networks.

Keywords: anomaly-based detection, cloud computing, intrusion detection, intrusion prevention, signature-based detection

Procedia PDF Downloads 307

Authors: Essang Anwana Onuntuei, Chinyere Blessing Azunwoke

Abstract:

The study used secondary data from foreign and local organizations to explore major challenges and opportunities abound in Information Communication. The study aimed at exploring the tie between tele density (network coverage area) and the number of network subscriptions, probing if the degree of consumer complaints varies significantly among network providers, and assessing if network subscriptions do significantly influence the sector’s GDP contribution. Methods used for data analysis include Pearson product-moment correlation and regression analysis, and the Analysis of Variance (ANOVA) as well. At a two-tailed test of 0.05 confidence level, the results of findings established about 85.6% of network subscriptions were explained by tele density (network coverage area), and the number of network subscriptions; Consumer Complaints’ degree varied significantly among network providers as 80.158291 (F calculated) > 3.490295 (F critical) with very high confidence associated p-value = 0.000000 which is < 0.05; and finally, 65% of the nation’s GDP was explained by network subscription to show a high association.

Keywords: tele density, subscription, network coverage, information communication, consumer

Procedia PDF Downloads 45

Authors: G. Anjan Babu, G. Sumana, M. Rajasekhar

Abstract:

Many inherited diseases and non-hereditary disorders are common in the development of renal cystic diseases. Polycystic kidney disease (PKD) is a disorder developed within the kidneys in which grouping of cysts filled with water like fluid. PKD is responsible for 5-10% of end-stage renal failure treated by dialysis or transplantation. New experimental models, application of molecular biology techniques have provided new insights into the pathogenesis of PKD. Researchers are showing keen interest for developing an automated system by applying computer aided techniques for the diagnosis of diseases. In this paper a multi-layered feed forward neural network with one hidden layer is constructed, trained and tested by applying back propagation learning rule for the diagnosis of PKD based on physical symptoms and test results of urinanalysis collected from the individual patients. The data collected from 50 patients are used to train and test the network. Among these samples, 75% of the data used for training and remaining 25% of the data are used for testing purpose. Furthermore, this trained network is used to implement for new samples. The output results in normality and abnormality of the patient.

Keywords: dialysis, hereditary, transplantation, polycystic, pathogenesis

Procedia PDF Downloads 380

Authors: Seedwell Sithole

Abstract:

One of the persistent problems in accounting education is how to effectively support students’ learning. A promising technique to this issue is to investigate the extent that learning is determined by the design of instructional material. This study examines the academic performance of students using three instructional designs in financial accounting. Student’s performance scores and reported mental effort ratings were used to determine the instructional effectiveness. The findings of this study show that accounting students prefer graph and text designs that are integrated. The results suggest that spatially separated graph and text presentations in accounting should be reorganized to align with the requirements of human cognitive architecture.

Keywords: accounting, cognitive load, education, instructional preferences, students

Procedia PDF Downloads 151

Authors: Kaustav Mukherjee

Abstract:

In this paper we propose a Neural Style Transfer solution whereby we have created a Lightweight Separable Convolution Kernel Based GAN Architecture (SC-GAN) which will very useful for designing filter for Mobile Phone Cameras and also Edge Devices which will convert any image to its 2D ANIMATED COMIC STYLE Movies like HEMAN, SUPERMAN, JUNGLE-BOOK. This will help the 2D animation artist by relieving to create new characters from real life person's images without having to go for endless hours of manual labour drawing each and every pose of a cartoon. It can even be used to create scenes from real life images.This will reduce a huge amount of turn around time to make 2D animated movies and decrease cost in terms of manpower and time. In addition to that being extreme light-weight it can be used as camera filters capable of taking Comic Style Shots using mobile phone camera or edge device cameras like Raspberry Pi 4,NVIDIA Jetson NANO etc. Existing Methods like CartoonGAN with the model size close to 170 MB is too heavy weight for mobile phones and edge devices due to their scarcity in resources. Compared to the current state of the art our proposed method which has a total model size of 31 MB which clearly makes it ideal and ultra-efficient for designing of camera filters on low resource devices like mobile phones, tablets and edge devices running OS or RTOS. .Owing to use of high resolution input and usage of bigger convolution kernel size it produces richer resolution Comic-Style Pictures implementation with 6 times lesser number of parameters and with just 25 extra epoch trained on a dataset of less than 1000 which breaks the myth that all GAN need mammoth amount of data. Our network reduces the density of the Gan architecture by using Depthwise Separable Convolution which does the convolution operation on each of the RGB channels separately then we use a Point-Wise Convolution to bring back the network into required channel number using 1 by 1 kernel.This reduces the number of parameters substantially and makes it extreme light-weight and suitable for mobile phones and edge devices. The architecture mentioned in the present paper make use of Parameterised Batch Normalization Goodfellow etc al. (Deep Learning OPTIMIZATION FOR TRAINING DEEP MODELS page 320) which makes the network to use the advantage of Batch Norm for easier training while maintaining the non-linear feature capture by inducing the learnable parameters

Keywords: comic stylisation from camera image using GAN, creating 2D animated movie style custom stickers from images, depth-wise separable convolutional neural network for light-weight GAN architecture for EDGE devices, GAN architecture for 2D animated cartoonizing neural style, neural style transfer for edge, model distilation, perceptual loss

Procedia PDF Downloads 132

Authors: Akinbo Razak Yinka, Adesanya Kehinde Kazeem, Oladokun Oluwagbenga Peter

Abstract:

Time series data are sequences of observations collected over a period of time. Such data can be used to predict health outcomes, such as disease progression, mortality, hospitalization, etc. The Statistical approach is based on mathematical models that capture the patterns and trends of the data, such as autocorrelation, seasonality, and noise, while Neural methods are based on artificial neural networks, which are computational models that mimic the structure and function of biological neurons. This paper compared both parametric and non-parametric time series models of patients treated for malaria in Maternal and Child Health Centres in Lagos State, Nigeria. The forecast methods considered linear regression, Integrated Moving Average, ARIMA and SARIMA Modeling for the parametric approach, while Multilayer Perceptron (MLP) and Long Short-Term Memory (LSTM) Network were used for the non-parametric model. The performance of each method is evaluated using the Mean Absolute Error (MAE), R-squared (R2) and Root Mean Square Error (RMSE) as criteria to determine the accuracy of each model. The study revealed that the best performance in terms of error was found in MLP, followed by the LSTM and ARIMA models. In addition, the Bootstrap Aggregating technique was used to make robust forecasts when there are uncertainties in the data.

Keywords: ARIMA, bootstrap aggregation, MLP, LSTM, SARIMA, time-series analysis

Procedia PDF Downloads 75

Authors: Hossein Havaeji, Tony Wong, Thien-My Dao

Abstract:

1. Research Problems and Research Objectives: Blockchain Technology-enabled Supply Chain System (BT-enabled SCS) is the system using BT to drive SCS transparency, security, durability, and process integrity as SCS data is not always visible, available, or trusted. The costs of operating BT in the SCS are a common problem in several organizations. The costs must be estimated as they can impact existing cost control strategies. To account for system and deployment costs, it is necessary to overcome the following hurdle. The problem is that the costs of developing and running a BT in SCS are not yet clear in most cases. Many industries aiming to use BT have special attention to the importance of BT installation cost which has a direct impact on the total costs of SCS. Predicting BT installation cost in SCS may help managers decide whether BT is to be an economic advantage. The purpose of the research is to identify some main BT installation cost components in SCS needed for deeper cost analysis. We then identify and categorize the main groups of cost components in more detail to utilize them in the prediction process. The second objective is to determine the suitable Supervised Learning technique in order to predict the costs of developing and running BT in SCS in a particular case study. The last aim is to investigate how the running BT cost can be involved in the total cost of SCS. 2. Work Performed: Applied successfully in various fields, Supervised Learning is a method to set the data frame, treat the data, and train/practice the method sort. It is a learning model directed to make predictions of an outcome measurement based on a set of unforeseen input data. The following steps must be conducted to search for the objectives of our subject. The first step is to make a literature review to identify the different cost components of BT installation in SCS. Based on the literature review, we should choose some Supervised Learning methods which are suitable for BT installation cost prediction in SCS. According to the literature review, some Supervised Learning algorithms which provide us with a powerful tool to classify BT installation components and predict BT installation cost are the Support Vector Regression (SVR) algorithm, Back Propagation (BP) neural network, and Artificial Neural Network (ANN). Choosing a case study to feed data into the models comes into the third step. Finally, we will propose the best predictive performance to find the minimum BT installation costs in SCS. 3. Expected Results and Conclusion: This study tends to propose a cost prediction of BT installation in SCS with the help of Supervised Learning algorithms. At first attempt, we will select a case study in the field of BT-enabled SCS, and then use some Supervised Learning algorithms to predict BT installation cost in SCS. We continue to find the best predictive performance for developing and running BT in SCS. Finally, the paper will be presented at the conference.

Keywords: blockchain technology, blockchain technology-enabled supply chain system, installation cost, supervised learning

Procedia PDF Downloads 122

Authors: Hazem J. Smadi

Abstract:

Performance of a supply chain is directly related to a distribution network that entails the location of storing materials or products and how products are delivered to the end customer through different stages in the supply chain. This study analyses the current distribution network used for delivering gas cylinders to end customer in Jordan. Evaluation of current distribution has been conducted across customer service components. A modification on the current distribution network in terms of central warehousing in each city in the country improves the response time and customer experience. 

Keywords: distribution network, gas cylinder, Jordan, supply chain

Procedia PDF Downloads 459

Authors: R. Poonkuzhali, M. Y. Sanavullah, M. R. Gurupriya

Abstract:

Opportunistic routing is used, where the network has the features like dynamic topology changes and intermittent network connectivity. In Delay Tolerant network or Disruption tolerant network opportunistic forwarding technique is widely used. The key idea of opportunistic routing is selecting forwarding nodes to forward data and coordination among these nodes to avoid duplicate transmissions. This paper gives the analysis of pros and cons of various opportunistic routing techniques used in MANET.

Keywords: ETX, opportunistic routing, PSR, throughput

Procedia PDF Downloads 494

Authors: Deepika Bhalla, Raj Kumar Bansal, Hari Om Gupta

Abstract:

Dissolved Gas Analysis (DGA) has been widely used for fault diagnosis in a transformer. Artificial neural networks (ANN) have high accuracy but are regarded as black boxes that are difficult to interpret. For many problems it is desired to extract knowledge from trained neural networks (NN) so that the user can gain a better understanding of the solution arrived by the NN. This paper applies a pedagogical approach for rule extraction from function approximating neural networks (REFANN) with application to incipient fault diagnosis using the concentrations of the dissolved gases within the transformer oil, as the input to the NN. The input space is split into subregions and for each subregion there is a linear equation that is used to predict the type of fault developing within a transformer. The experiments on real data indicate that the approach used can extract simple and useful rules and give fault predictions that match the actual fault and are at times also better than those predicted by the IEC method.

Keywords: artificial neural networks, dissolved gas analysis, rules extraction, transformer

Procedia PDF Downloads 536

Authors: Zhong Ma, Zhuping Wang, Congxin Liu, Xiangzeng Liu

Abstract:

Satellite imagery classification is a challenging problem with many practical applications. In this paper, we designed a deep convolution neural network (DCNN) to classify the satellite imagery. The contributions of this paper are twofold — First, to cope with the large-scale variance in the satellite image, we introduced the inception module, which has multiple filters with different size at the same level, as the building block to build our DCNN model. Second, we proposed a genetic algorithm based method to efficiently search the best hyper-parameters of the DCNN in a large search space. The proposed method is evaluated on the benchmark database. The results of the proposed hyper-parameters search method show it will guide the search towards better regions of the parameter space. Based on the found hyper-parameters, we built our DCNN models, and evaluated its performance on satellite imagery classification, the results show the classification accuracy of proposed models outperform the state of the art method.

Keywords: satellite imagery classification, deep convolution network, genetic algorithm, hyper-parameter optimization

Procedia PDF Downloads 301

Authors: Kwok W. Mui, Ling T. Wong, Chi T. Cheung

Abstract:

Cooling energy consumption in the residential sector, different from shopping mall, office or commercial buildings, is significantly subject to occupant decisions where in-depth investigations are found limited. It shows that energy consumptions could be associated with housing types. Surveys have been conducted in existing Hong Kong public housings to understand the housing characteristics, apartment electricity demands, occupant’s thermal expectations, and air–conditioning usage patterns for further cooling energy-saving assessments. The aim of this study is to develop a hybrid cooling energy prediction model, which integrated by EnergyPlus (EP) and artificial neural network (ANN) to estimate cooling energy consumption in public residential sector. Sensitivity tests are conducted to find out the energy impacts with changing building parameters regarding to external wall and window material selection, window size reduction, shading extension, building orientation and apartment size control respectively. Assessments are performed to investigate the relationships between cooling demands and occupant behavior on thermal environment criteria and air-conditioning operation patterns. The results are summarized into a cooling energy calculator for layman use to enhance the cooling energy saving awareness in their own living environment. The findings can be used as a directory framework for future cooling energy evaluation in residential buildings, especially focus on the occupant behavioral air–conditioning operation and criteria of energy-saving incentives.

Keywords: artificial neural network, cooling energy, occupant behavior, residential buildings, thermal environment

Procedia PDF Downloads 168

Authors: Tayo P. Ogundunmade, Adedayo A. Adepoju

Abstract:

Terrorist attacks in liberal democracies bring about a few pessimistic results, for example, sabotaged public support in the governments they target, disturbing the peace of a protected environment underwritten by the state, and a limitation of individuals from adding to the advancement of the country, among others. Hence, seeking for techniques to understand the different factors involved in terrorism and how to deal with those factors in order to completely stop or reduce terrorist activities is the topmost priority of the government in every country. This research aim is to develop an efficient deep learning-based predictive model for the prediction of future terrorist activities in Nigeria, addressing low-quality prediction accuracy problems associated with the existing solution methods. The proposed predictive AI-based model as a counterterrorism tool will be useful by governments and law enforcement agencies to protect the lives of individuals in society and to improve the quality of life in general. A Heterogeneous Bayesian Neural Network (HETBNN) model was derived with Gaussian error normal distribution. Three primary transfer functions (HOTTFs), as well as two derived transfer functions (HETTFs) arising from the convolution of the HOTTFs, are namely; Symmetric Saturated Linear transfer function (SATLINS ), Hyperbolic Tangent transfer function (TANH), Hyperbolic Tangent sigmoid transfer function (TANSIG), Symmetric Saturated Linear and Hyperbolic Tangent transfer function (SATLINS-TANH) and Symmetric Saturated Linear and Hyperbolic Tangent Sigmoid transfer function (SATLINS-TANSIG). Data on the Terrorist activities in Nigeria gathered through questionnaires for the purpose of this study were used. Mean Square Error (MSE), Mean Absolute Error (MAE) and Test Error are the forecast prediction criteria. The results showed that the HETFs performed better in terms of prediction and factors associated with terrorist activities in Nigeria were determined. The proposed predictive deep learning-based model will be useful to governments and law enforcement agencies as an effective counterterrorism mechanism to understand the parameters of terrorism and to design strategies to deal with terrorism before an incident actually happens and potentially causes the loss of precious lives. The proposed predictive AI-based model will reduce the chances of terrorist activities and is particularly helpful for security agencies to predict future terrorist activities.

Keywords: activation functions, Bayesian neural network, mean square error, test error, terrorism

Procedia PDF Downloads 165

Authors: P. Suresh, K. Gunasekaran, R. Thanigaivelan

Abstract:

Radio frequency identification (RFID) technology has been attracting considerable attention with the expectation of improved supply chain visibility for consumer goods, apparel, and pharmaceutical manufacturers, as well as retailers and government procurement agencies. It is also expected to improve the consumer shopping experience by making it more likely that the products they want to purchase are available. Recent announcements from some key retailers have brought interest in RFID to the forefront. A modified K- Means Cluster based Heuristic approach, Hybrid Genetic Algorithm (GA) - Simulated Annealing (SA) approach, Hybrid K-Means Cluster based Heuristic-GA and Hybrid K-Means Cluster based Heuristic-GA-SA for Open Loop Supply Chain Network problem are proposed. The study incorporated uniform crossover operator and combined crossover operator in GAs for solving open loop supply chain distribution network problem. The algorithms are tested on 50 randomly generated data set and compared with each other. The results of the numerical experiments show that the Hybrid K-means cluster based heuristic-GA-SA, when tested on 50 randomly generated data set, shows superior performance to the other methods for solving the open loop supply chain distribution network problem.

Keywords: RFID, supply chain distribution network, open loop supply chain, genetic algorithm, simulated annealing

Procedia PDF Downloads 165

Authors: Ava Salmanpour, Hanieh Saeedi, Payam Rouhi, Elahe Hamzeil, Shima Alimohammadi, Siamak Hossein Khalaj, Mohammad Asadian

Abstract:

Software-defined network (SDN) is a network architecture designed to control network using software application in a central manner. This ability enables remote control of the whole network regardless of the network technology. In fact, in this architecture network intelligence is separated from physical infrastructure, it means that required network components can be implemented virtually using software applications. Today, power networks are characterized by a high range of complexity with a large number of intelligent devices, processing both huge amounts of data and important information. Therefore, reliable and secure communication networks are required. SDNs are the best choice to meet this issue. In this paper, SDN networks capabilities and characteristics will be reviewed and different basic controllers will be compared. The importance of using SDNs to escalate efficiency and reliability in utility power networks is going to be discussed and the comparison between the SDN-based power networks and traditional networks will be explained.

Keywords: software-defined network, SDNs, utility network, open flow, communication, gas and electricity, controller

Procedia PDF Downloads 113

Authors: Milad Abbasi

Abstract:

Carbon fiber-reinforced polymer (CFRP) composite structures are increasingly being utilized in the automotive industry due to their lightweight and specific energy absorption capabilities. Although it is impossible to predict composite mechanical properties directly using theoretical methods, various research has been conducted so far in the literature for accurate simulation of CFRP structures' energy-absorbing behavior. In this research, axial compression experiments were carried out on hand lay-up unidirectional CFRP composite tubes. The fabrication method allowed the authors to extract the material properties of the CFRPs using ASTM D3039, D3410, and D3518 standards. A neural network machine learning algorithm was then utilized to build a robust prediction model to forecast the axial compressive properties of CFRP tubes while reducing high-cost experimental efforts. The predicted results have been compared with the experimental outcomes in terms of load-carrying capacity and energy absorption capability. The results showed high accuracy and precision in the prediction of the energy-absorption capacity of the CFRP tubes. This research also demonstrates the effectiveness and challenges of machine learning techniques in the robust simulation of composites' energy-absorption behavior. Interestingly, the proposed method considerably condensed numerical and experimental efforts in the simulation and calibration of CFRP composite tubes subjected to compressive loading.

Keywords: CFRP composite tubes, energy absorption, crushing behavior, machine learning, neural network

Procedia PDF Downloads 153

Authors: Zhifeng Kong

Abstract:

Over-parameterized neural networks have attracted a great deal of attention in recent deep learning theory research, as they challenge the classic perspective of over-fitting when the model has excessive parameters and have gained empirical success in various settings. While a number of theoretical works have been presented to demystify properties of such models, the convergence properties of such models are still far from being thoroughly understood. In this work, we study the convergence properties of training two-hidden-layer partially over-parameterized fully connected networks with the Rectified Linear Unit activation via gradient descent. To our knowledge, this is the first theoretical work to understand convergence properties of deep over-parameterized networks without the equally-wide-hidden-layer assumption and other unrealistic assumptions. We provide a probabilistic lower bound of the widths of hidden layers and proved linear convergence rate of gradient descent. We also conducted experiments on synthetic and real-world datasets to validate our theory.

Keywords: over-parameterization, rectified linear units ReLU, convergence, gradient descent, neural networks

Procedia PDF Downloads 142