Search results for: artificial neural network modeling

Authors: Vijaya Yuvaram Singh V M, Kameshwar Rao J V

Abstract:

The challenge with development of neural network based methods for medical is the availability of data. Anatomical landmark detection in the medical domain is a process to find points on the x-ray scan report of the patient. Most of the time this task is done manually by trained professionals as it requires precision and domain knowledge. Traditionally object detection based methods are used for landmark detection. Here, we utilize reinforcement learning and query based method to train a single agent capable of detecting multiple landmarks. A deep Q network agent is trained to detect single and multiple landmarks present on hip and shoulder from x-ray scan of a patient. Here a single agent is trained to find multiple landmark making it superior to having individual agents per landmark. For the initial study, five images of different patients are used as the environment and tested the agents performance on two unseen images.

Keywords: reinforcement learning, medical landmark detection, multi target detection, deep neural network

Procedia PDF Downloads 142

Authors: Yakubu Bala Mohammed, Nadire Chavus, Mohammed Bulama

Abstract:

Generative Pre-trained Transformer (ChatGPT) represents an artificial intelligence (AI) tool capable of swiftly generating comprehensive responses to prompts and follow-up inquiries. This emerging AI tool was introduced in November 2022 by OpenAI firm, an American AI research laboratory, utilizing substantial language models. This present study aims to delve into the potential future consequences of ChatGPT usage in education using AI-based algorithms. The paper will bring forth the likely potential risks of ChatGBT utilization, such as academic integrity concerns, unfair learning assessments, excessive reliance on AI, and dissemination of inaccurate information using four machine learning algorithms: eXtreme-Gradient Boosting (XGBoost), Support vector machine (SVM), Emotional artificial neural network (EANN), and Random forest (RF) would be used to analyze the study collected data due to their robustness. Finally, the findings of the study will assist education stakeholders in understanding the future implications of ChatGPT usage in education and propose solutions and directions for upcoming studies.

Keywords: machine learning, ChatGPT, education, learning, implications

Procedia PDF Downloads 231

Authors: Zhihao Zheng, Zhilin Wang, Linxin Liu

Abstract:

In the research of automated knowledge question-answering systems, accuracy and efficiency are critical challenges. This paper proposes a knowledge graph-enhanced Retrieval-Augmented Generation (RAG) method, combined with a Graph Neural Network (GNN) structure, to automatically determine the correctness of knowledge competition questions. First, a domain-specific knowledge graph was constructed from a large corpus of academic journal literature, with key entities and relationships extracted using Natural Language Processing (NLP) techniques. Then, the RAG method's retrieval module was expanded to simultaneously query both text databases and the knowledge graph, leveraging the GNN to further extract structured information from the knowledge graph. During answer generation, contextual information provided by the knowledge graph and GNN is incorporated to improve the accuracy and consistency of the answers. Experimental results demonstrate that the knowledge graph and GNN-enhanced RAG method perform excellently in determining the correctness of questions, achieving an accuracy rate of 95%. Particularly in cases involving ambiguity or requiring contextual information, the structured knowledge provided by the knowledge graph and GNN significantly enhances the RAG method's performance. This approach not only demonstrates significant advantages in improving the accuracy and efficiency of automated knowledge question-answering systems but also offers new directions and ideas for future research and practical applications.

Keywords: knowledge graph, graph neural network, retrieval-augmented generation, NLP

Procedia PDF Downloads 39

Authors: Sofia Stoica

Abstract:

In this paper, we use historical data to predict the stock price of a tech company. To this end, we use a dataset consisting of the stock prices in the past five years of ten major tech companies – Adobe, Amazon, Apple, Facebook, Google, Microsoft, Netflix, Oracle, Salesforce, and Tesla. We experimented with a variety of models– a linear regressor model, K nearest Neighbors (KNN), a sequential neural network – and algorithms - Multiplicative Weight Update, and AdaBoost. We found that the sequential neural network performed the best, with a testing error of 0.18%. Interestingly, the linear model performed the second best with a testing error of 0.73%. These results show that using historical data is enough to obtain high accuracies, and a simple algorithm like linear regression has a performance similar to more sophisticated models while taking less time and resources to implement.

Keywords: finance, machine learning, opening price, stock market

Procedia PDF Downloads 189

Authors: David P. Tantua, Abdulkader Helwan

Abstract:

Biometrics has been recently used for the human identification systems using the biological traits such as the fingerprints and iris scanning. Identification systems based biometrics show great efficiency and accuracy in such human identification applications. However, these types of systems are so far based on some image processing techniques only, which may decrease the efficiency of such applications. Thus, this paper aims to develop a human palmprint identification system using multi-layer perceptron neural network which has the capability to learn using a backpropagation learning algorithms. The developed system uses images obtained from a public database available on the internet (CASIA). The processing system is as follows: image filtering using median filter, image adjustment, image skeletonizing, edge detection using canny operator to extract features, clear unwanted components of the image. The second phase is to feed those processed images into a neural network classifier which will adaptively learn and create a class for each different image. 100 different images are used for training the system. Since this is an identification system, it should be tested with the same images. Therefore, the same 100 images are used for testing it, and any image out of the training set should be unrecognized. The experimental results shows that this developed system has a great accuracy 100% and it can be implemented in real life applications.

Keywords: biometrics, biological traits, multi-layer perceptron neural network, image skeletonizing, edge detection using canny operator

Procedia PDF Downloads 371

Authors: Fengjuan Wang, Azilawati Jamaludin

Abstract:

Developmental Dyscalculia (DD) is defined as a particular learning difficulty with continuous challenges in learning requisite math skills that cannot be explained by intellectual disability or educational deprivation. Recent studies have increasingly recognized that DD is a heterogeneous, instead of monolithic, learning disorder with not only cognitive and behavioral deficits but so too neural dysfunction. In recent years, neuroimaging studies employed group comparison to explore the neural underpinnings of DD, which contradicted the heterogenous nature of DD and may obfuscate critical individual differences. This research aimed to investigate the neural heterogeneity of DD using case studies with functional near-infrared spectroscopy (fNIRS). A total of 54 aged 6-7 years old of children participated in this study, comprising two comprehensive cognitive assessments, an 8-minute resting state, and an 8-minute one-digit addition task. Nine children met the criteria of DD and scored at or below 85 (i.e., the 16th percentile) on the Mathematics or Math Fluency subtest of the Wechsler Individual Achievement Test, Third Edition (WIAT-III) (both subtest scores were 90 and below). The remaining 45 children formed the typically developing (TD) group. Resting-state data and brain activation in the inferior frontal gyrus (IFG), superior frontal gyrus (SFG), and intraparietal sulcus (IPS) were collected for comparison between each case and the TD group. Graph theory was used to analyze the brain network under the resting state. This theory represents the brain network as a set of nodes--brain regions—and edges—pairwise interactions across areas to reveal the architectural organizations of the nervous network. Next, a single-case methodology developed by Crawford et al. in 2010 was used to compare each case’s brain network indicators and brain activation against 45 TD children’s average data. Results showed that three out of the nine DD children displayed significant deviation from TD children’s brain indicators. Case 1 had inefficient nodal network properties. Case 2 showed inefficient brain network properties and weaker activation in the IFG and IPS areas. Case 3 displayed inefficient brain network properties with no differences in activation patterns. As a rise above, the present study was able to distill differences in architectural organizations and brain activation of DD vis-à-vis TD children using fNIRS and single-case methodology. Although DD is regarded as a heterogeneous learning difficulty, it is noted that all three cases showed lower nodal efficiency in the brain network, which may be one of the neural sources of DD. Importantly, although the current “brain norm” established for the 45 children is tentative, the results from this study provide insights not only for future work in “developmental brain norm” with reliable brain indicators but so too the viability of single-case methodology, which could be used to detect differential brain indicators of DD children for early detection and interventions.

Keywords: brain activation, brain network, case study, developmental dyscalculia, functional near-infrared spectroscopy, graph theory, neural heterogeneity

Procedia PDF Downloads 53

Authors: Aleksei Samarin, Artyom Nazarenko, Valentin Malykh

Abstract:

Image enhancement is becoming more and more popular, especially on mobile devices. Nowadays, it is a common approach to enhance an image using a convolutional neural network (CNN). Such a network should be of significant size; otherwise, a possibility for the artifacts to occur is overgrowing. The existing large CNNs are computationally expensive, which could be crucial for mobile devices. Another important flaw of such models is they are poorly interpretable. There is another approach to image enhancement, namely, the usage of predefined filters in combination with the prediction of their applicability. We present an approach following this paradigm, which outperforms both existing CNN-based and filter-based approaches in the image enhancement task. It is easily adaptable for mobile devices since it has only 47 thousand parameters. It shows the best SSIM 0.919 on RANDOM250 (MIT Adobe FiveK) among small models and is thrice faster than previous models.

Keywords: universal filter, image enhancement, neural networks, computer vision

Procedia PDF Downloads 101

Authors: Zekang Lan, Yan Xu, Yingkun Huang, Dian Huang, Shengzhong Feng

Abstract:

Jobs on high-performance computing (HPC) clusters can suffer significant performance degradation due to inter-job network interference. Topology-aware job allocation problem (TJAP) is such a problem that decides how to dedicate nodes to specific applications to mitigate inter-job network interference. In this paper, we study the window-based TJAP on a fat-tree network aiming at minimizing the cost of communication hop, a defined inter-job interference metric. The window-based approach for scheduling repeats periodically, taking the jobs in the queue and solving an assignment problem that maps jobs to the available nodes. Two special allocation strategies are considered, i.e., static continuity assignment strategy (SCAS) and dynamic continuity assignment strategy (DCAS). For the SCAS, a 0-1 integer programming is developed. For the DCAS, an approach called neural simulated algorithm (NSA), which is an extension to simulated algorithm (SA) that learns a repair operator and employs them in a guided heuristic search, is proposed. The efficacy of NSA is demonstrated with a computational study against SA and SCIP. The results of numerical experiments indicate that both the model and algorithm proposed in this paper are effective.

Keywords: high-performance computing, job allocation, neural simulated annealing, topology-aware

Procedia PDF Downloads 116

Authors: Brwa Abdulrahman Abubaker

Abstract:

Recently, there has been a lot of interest in the difficult task of applying reinforcement learning to autonomous mobile robots. Conventional reinforcement learning (TRL) techniques have many drawbacks, such as lengthy computation times, intricate control frameworks, a great deal of trial and error searching, and sluggish convergence. In this paper, a modified Spiking Neural Network (SNN) is used to offer a distinct method for autonomous mobile robot learning and control in unexpected surroundings. As a learning algorithm, the suggested model combines dopamine modulation with spike-timing-dependent plasticity (STDP). In order to create more computationally efficient, biologically inspired control systems that are adaptable to changing settings, this work uses the effective and physiologically credible Izhikevich neuron model. This study is primarily focused on creating an algorithm for target tracking in the presence of obstacles. Results show that the SNN trained with three obstacles yielded an impressive 96% success rate for our proposal, with collisions happening in about 4% of the 214 simulated seconds.

Keywords: spiking neural network, spike-timing-dependent plasticity, dopamine modulation, reinforcement learning

Procedia PDF Downloads 21

Authors: Ana Neves, John Leander, Ignacio Gonzalez, Raid Karoumi

Abstract:

Systematic monitoring and inspection are needed to assess the present state of a structure and predict its future condition. If an irregularity is noticed, repair actions may take place and the adequate intervention will most probably reduce the future costs with maintenance, minimize downtime and increase safety by avoiding the failure of the structure as a whole or of one of its structural parts. For this to be possible decisions must be made at the right time, which implies using systems that can detect abnormalities in their early stage. In this sense, Structural Health Monitoring (SHM) is seen as an effective tool for improving the safety and reliability of infrastructures. This paper explores the decision-making problem in SHM regarding the maintenance of civil engineering structures. The aim is to assess the present condition of a bridge based exclusively on measurements using the suggested method in this paper, such that action is taken coherently with the information made available by the monitoring system. Artificial Neural Networks are trained and their ability to predict structural behavior is evaluated in the light of a case study where acceleration measurements are acquired from a bridge located in Stockholm, Sweden. This relatively old bridge is presently still in operation despite experiencing obvious problems already reported in previous inspections. The prediction errors provide a measure of the accuracy of the algorithm and are subjected to further investigation, which comprises concepts like clustering analysis and statistical hypothesis testing. These enable to interpret the obtained prediction errors, draw conclusions about the state of the structure and thus support decision making regarding its maintenance.

Keywords: artificial neural networks, clustering analysis, model-free damage detection, statistical hypothesis testing, structural health monitoring

Procedia PDF Downloads 208

Authors: Luis Amador Jimenez, Md. Shohel Amin

Abstract:

Pavement performance models are based on projections of observed traffic loads, which makes uncertain to study funding strategies in the long run if history does not repeat. Neural networks can be used to estimate deterioration rates but the learning rate and momentum have not been properly investigated, in addition, economic evolvement could change traffic flows. This study addresses both issues through a case study for roads of Montreal that simulates traffic for a period of 50 years and deals with the measurement error of the pavement deterioration model. Travel demand models are applied to simulate annual average daily traffic (AADT) every 5 years. Accumulated equivalent single axle loads (ESALs) are calculated from the predicted AADT and locally observed truck distributions combined with truck factors. A back propagation Neural Network (BPN) method with a Generalized Delta Rule (GDR) learning algorithm is applied to estimate pavement deterioration models capable of overcoming measurement errors. Linear programming of lifecycle optimization is applied to identify M&R strategies that ensure good pavement condition while minimizing the budget. It was found that CAD 150 million is the minimum annual budget to good condition for arterial and local roads in Montreal. Montreal drivers prefer the use of public transportation for work and education purposes. Vehicle traffic is expected to double within 50 years, ESALS are expected to double the number of ESALs every 15 years. Roads in the island of Montreal need to undergo a stabilization period for about 25 years, a steady state seems to be reached after.

Keywords: pavement management system, traffic simulation, backpropagation neural network, performance modeling, measurement errors, linear programming, lifecycle optimization

Procedia PDF Downloads 460

Authors: Abdulmalek Ahmed

Abstract:

Formation pressure is the main function that affects drilling operation economically and efficiently. Knowing the pore pressure and the parameters that affect it will help to reduce the cost of drilling process. Many empirical models reported in the literature were used to calculate the formation pressure based on different parameters. Some of these models used only drilling parameters to estimate pore pressure. Other models predicted the formation pressure based on log data. All of these models required different trends such as normal or abnormal to predict the pore pressure. Few researchers applied artificial intelligence (AI) techniques to predict the formation pressure by only one method or a maximum of two methods of AI. The objective of this research is to predict the pore pressure based on both drilling parameters and log data namely; weight on bit, rotary speed, rate of penetration, mud weight, bulk density, porosity and delta sonic time. A real field data is used to predict the formation pressure using five different artificial intelligence (AI) methods such as; artificial neural networks (ANN), radial basis function (RBF), fuzzy logic (FL), support vector machine (SVM) and functional networks (FN). All AI tools were compared with different empirical models. AI methods estimated the formation pressure by a high accuracy (high correlation coefficient and low average absolute percentage error) and outperformed all previous. The advantage of the new technique is its simplicity, which represented from its estimation of pore pressure without the need of different trends as compared to other models which require a two different trend (normal or abnormal pressure). Moreover, by comparing the AI tools with each other, the results indicate that SVM has the advantage of pore pressure prediction by its fast processing speed and high performance (a high correlation coefficient of 0.997 and a low average absolute percentage error of 0.14%). In the end, a new empirical correlation for formation pressure was developed using ANN method that can estimate pore pressure with a high precision (correlation coefficient of 0.998 and average absolute percentage error of 0.17%).

Keywords: Artificial Intelligence (AI), Formation pressure, Artificial Neural Networks (ANN), Fuzzy Logic (FL), Support Vector Machine (SVM), Functional Networks (FN), Radial Basis Function (RBF)

Procedia PDF Downloads 149

Authors: Jonathan Sohn

Abstract:

Deep neural networks (DNNs) have shown state-of-the-art performance for many applications, including computer vision, natural language processing, and speech recognition. Recently, adversarial attacks have been studied in the context of deep neural networks, which aim to alter the results of deep neural networks by modifying the inputs slightly. For example, an adversarial attack on a DNN used for object detection can cause the DNN to miss certain objects. As a result, the reliability of DNNs is undermined by their lack of robustness against adversarial attacks, raising concerns about their use in safety-critical applications such as autonomous driving. In this paper, we focus on studying the adversarial attacks and defenses on DNNs for image classification. There are two types of adversarial attacks studied which are fast gradient sign method (FGSM) attack and projected gradient descent (PGD) attack. A DNN forms decision boundaries that separate the input images into different categories. The adversarial attack slightly alters the image to move over the decision boundary, causing the DNN to misclassify the image. FGSM attack obtains the gradient with respect to the image and updates the image once based on the gradients to cross the decision boundary. PGD attack, instead of taking one big step, repeatedly modifies the input image with multiple small steps. There is also another type of attack called the target attack. This adversarial attack is designed to make the machine classify an image to a class chosen by the attacker. We can defend against adversarial attacks by incorporating adversarial examples in training. Specifically, instead of training the neural network with clean examples, we can explicitly let the neural network learn from the adversarial examples. In our experiments, the digit recognition accuracy on the MNIST dataset drops from 97.81% to 39.50% and 34.01% when the DNN is attacked by FGSM and PGD attacks, respectively. If we utilize FGSM training as a defense method, the classification accuracy greatly improves from 39.50% to 92.31% for FGSM attacks and from 34.01% to 75.63% for PGD attacks. To further improve the classification accuracy under adversarial attacks, we can also use a stronger PGD training method. PGD training improves the accuracy by 2.7% under FGSM attacks and 18.4% under PGD attacks over FGSM training. It is worth mentioning that both FGSM and PGD training do not affect the accuracy of clean images. In summary, we find that PGD attacks can greatly degrade the performance of DNNs, and PGD training is a very effective way to defend against such attacks. PGD attacks and defence are overall significantly more effective than FGSM methods.

Keywords: deep neural network, adversarial attack, adversarial defense, adversarial machine learning

Procedia PDF Downloads 194

Authors: Lizhi Ma, Dan Liu

Abstract:

Pipeline parallelism has been widely used to accelerate distributed deep learning to alleviate GPU memory bottlenecks and to ensure that models can be trained and deployed smoothly under limited graphics memory conditions. However, in highly heterogeneous distributed clusters, traditional model partitioning methods are not able to achieve load balancing. The overlap of communication and computation is also a big challenge. In this paper, HePipe is proposed, an efficient pipeline parallel training method for highly heterogeneous clusters. According to the characteristics of the neural network model pipeline training task, oriented to the 2-level heterogeneous cluster computing topology, a training method based on the 2-level stage division of neural network modeling and partitioning is designed to improve the parallelism. Additionally, a multi-forward 1F1B scheduling strategy is designed to accelerate the training time of each stage by executing the computation units in advance to maximize the overlap between the forward propagation communication and backward propagation computation. Finally, a dynamic recomputation strategy based on task memory requirement prediction is proposed to improve the fitness ratio of task and memory, which improves the throughput of the cluster and solves the memory shortfall problem caused by memory differences in heterogeneous clusters. The empirical results show that HePipe improves the training speed by 1.6×−2.2× over the existing asynchronous pipeline baselines.

Keywords: pipeline parallelism, heterogeneous cluster, model training, 2-level stage partitioning

Procedia PDF Downloads 18

Authors: Afaf Alharbi, Qianni Zhang

Abstract:

The identification of malignant tissue in histopathological slides holds significant importance in both clinical settings and pathology research. This paper introduces a methodology aimed at automatically categorizing cancerous tissue through the utilization of a multiple-instance learning framework. This framework is specifically developed to acquire knowledge of the Bernoulli distribution of the bag label probability by employing neural networks. Furthermore, we put forward a neural network based permutation-invariant aggregation operator, equivalent to attention mechanisms, which is applied to the multi-instance learning network. Through empirical evaluation of an openly available colon cancer histopathology dataset, we provide evidence that our approach surpasses various conventional deep learning methods.

Keywords: attention multiple instance learning, MIL and transfer learning, histopathological slides, cancer tissue classification

Procedia PDF Downloads 110

Authors: Neveen AlQasas, Daniel Johnson

Abstract:

Membranes for water treatment are an established technology that attracts great attention due to its simplicity and cost effectiveness. However, membranes in operation suffer from the adverse effect of membrane fouling. Bio-fouling is a phenomenon that occurs at the water-membrane interface, and is a dynamic process that is initiated by the adsorption of dissolved organic material, including biomacromolecules, on the membrane surface. After initiation, attachment of microorganisms occurs, followed by biofilm growth. The biofilm blocks the pores of the membrane and consequently results in reducing the water flux. Moreover, the presence of a fouling layer can have a substantial impact on the membrane separation properties. Understanding the mechanism of the initiation phase of biofouling is a key point in eliminating the biofouling on membrane surfaces. The adhesion and attachment of different fouling materials is affected by the surface properties of the membrane materials. Therefore, surface properties of different polymeric materials had been studied in terms of their surface energies and Hansen solubility parameters (HSP). The difference between the combined HSP parameters (HSP distance) allows prediction of the affinity of two materials to each other. The possibilities of measuring the HSP of different polymer films via surface measurements, such as contact angle has been thoroughly investigated. Knowing the HSP of a membrane material and the HSP of a specific foulant, facilitate the estimation of the HSP distance between the two, and therefore the strength of attachment to the surface. Contact angle measurements using fourteen different solvents on five different polymeric films were carried out using the sessile drop method. Solvents were ranked as good or bad solvents using different ranking method and ranking was used to calculate the HSP of each polymeric film. Results clearly indicate the absence of a direct relation between contact angle values of each film and the HSP distance between each polymer film and the solvents used. Therefore, estimating HSP via contact angle alone is not sufficient. However, it was found if the surface tensions and viscosities of the used solvents are taken in to the account in the analysis of the contact angle values, a prediction of the HSP from contact angle measurements is possible. This was carried out via training of a neural network model. The trained neural network model has three inputs, contact angle value, surface tension and viscosity of solvent used. The model is able to predict the HSP distance between the used solvent and the tested polymer (material). The HSP distance prediction is further used to estimate the total and individual HSP parameters of each tested material. The results showed an accuracy of about 90% for all the five studied films

Keywords: surface characterization, hansen solubility parameter estimation, contact angle measurements, artificial neural network model, surface measurements

Procedia PDF Downloads 94

Authors: Avinash Malladhi

Abstract:

Forensic accounting is a specialized field that involves the application of accounting principles, investigative skills, and legal knowledge to detect and prevent fraud. With the rise of big data and technological advancements, artificial intelligence (AI) and machine learning (ML) algorithms have emerged as powerful tools for forensic accountants to enhance their fraud detection capabilities. In this paper, we review and analyze various AI/ML algorithms that are commonly used in forensic accounting, including supervised and unsupervised learning, deep learning, natural language processing Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Support Vector Machines (SVMs), Decision Trees, and Random Forests. We discuss their underlying principles, strengths, and limitations and provide empirical evidence from existing research studies demonstrating their effectiveness in detecting financial fraud. We also highlight potential ethical considerations and challenges associated with using AI/ML in forensic accounting. Furthermore, we highlight the benefits of these technologies in improving fraud detection and prevention in forensic accounting.

Keywords: AI, machine learning, forensic accounting & fraud detection, anti money laundering, Benford's law, fraud triangle theory

Procedia PDF Downloads 93

Authors: Duncan Wallace, M-Tahar Kechadi

Abstract:

In recent years, Machine Learning (ML) approaches have been successfully applied to an analysis of patient symptom data in the context of disease diagnosis, at least where such data is well codified. However, much of the data present in Electronic Health Records (EHR) are unlikely to prove suitable for classic ML approaches. Furthermore, as scores of data are widely spread across both hospitals and individuals, a decentralized, computationally scalable methodology is a priority. The focus of this paper is to develop a method to predict outliers in an out-of-hours healthcare provision center (OOHC). In particular, our research is based upon the early identification of patients who have underlying conditions which will cause them to repeatedly require medical attention. OOHC act as an ad-hoc delivery of triage and treatment, where interactions occur without recourse to a full medical history of the patient in question. Medical histories, relating to patients contacting an OOHC, may reside in several distinct EHR systems in multiple hospitals or surgeries, which are unavailable to the OOHC in question. As such, although a local solution is optimal for this problem, it follows that the data under investigation is incomplete, heterogeneous, and comprised mostly of noisy textual notes compiled during routine OOHC activities. Through the use of Deep Learning methodologies, the aim of this paper is to provide the means to identify patient cases, upon initial contact, which are likely to relate to such outliers. To this end, we compare the performance of Long Short-Term Memory, Gated Recurrent Units, and combinations of both with Convolutional Neural Networks. A further aim of this paper is to elucidate the discovery of such outliers by examining the exact terms which provide a strong indication of positive and negative case entries. While free-text is the principal data extracted from EHRs for classification, EHRs also contain normalized features. Although the specific demographical features treated within our corpus are relatively limited in scope, we examine whether it is beneficial to include such features among the inputs to our neural network, or whether these features are more successfully exploited in conjunction with a different form of a classifier. In this section, we compare the performance of randomly generated regression trees and support vector machines and determine the extent to which our classification program can be improved upon by using either of these machine learning approaches in conjunction with the output of our Recurrent Neural Network application. The output of our neural network is also used to help determine the most significant lexemes present within the corpus for determining high-risk patients. By combining the confidence of our classification program in relation to lexemes within true positive and true negative cases, with an inverse document frequency of the lexemes related to these cases, we can determine what features act as the primary indicators of frequent-attender and non-frequent-attender cases, providing a human interpretable appreciation of how our program classifies cases.

Keywords: artificial neural networks, data-mining, machine learning, medical informatics

Procedia PDF Downloads 131

Authors: Zhaojun Wang, Zongdi Sun, Zhiyuan Liu

Abstract:

First of all, the carbon trading price and trading volume in Shanghai are transformed by Fourier transform, and the frequency response diagram is obtained. Then, the frequency response diagram is analyzed and the Blackman filter is designed. The Blackman filter is used to filter, and the carbon trading time domain and frequency response diagram are obtained. After wavelet analysis, the carbon trading data were processed; respectively, we got the average value for each 5 days, 10 days, 20 days, 30 days, and 60 days. Finally, the data are used as input of the Back Propagation Neural Network model for prediction.

Keywords: Shanghai carbon trading, carbon trading price, carbon trading volume, wavelet analysis, BP neural network model

Procedia PDF Downloads 391

Authors: Hervice Roméo Fogno Fotsoa, Germaine Djuidje Kenmoe, Claude Vidal Aloyem Kazé

Abstract:

One of the main applications of machine learning is the prediction of time series. But a more accurate prediction requires a more optimal model of machine learning. Several optimization techniques have been developed, but without considering the input variables disposition of the system. Thus, this work aims to present a new machine learning architecture optimization technique based on their optimal input variables disposition. The validations are done on the prediction of wind time series, using data collected in Cameroon. The number of possible dispositions with four input variables is determined, i.e., twenty-four. Each of the dispositions is used to perform the prediction, with the main criteria being the training and prediction performances. The results obtained from a static architecture and a dynamic architecture of neural networks have shown that these performances are a function of the input variable's disposition, and this is in a different way from the architectures. This analysis revealed that it is necessary to take into account the input variable's disposition for the development of a more optimal neural network model. Thus, a new neural network training algorithm is proposed by introducing the search for the optimal input variables disposition in the traditional back-propagation algorithm. The results of the application of this new optimization approach on the two single neural network architectures are compared with the previously obtained results step by step. Moreover, this proposed approach is validated in a collaborative optimization method with a single objective optimization technique, i.e., genetic algorithm back-propagation neural networks. From these comparisons, it is concluded that each proposed model outperforms its traditional model in terms of training and prediction performance of time series. Thus the proposed optimization approach can be useful in improving the accuracy of time series forecasts. This proves that the proposed optimization approach can be useful in improving the accuracy of time series prediction based on machine learning.

Keywords: input variable disposition, machine learning, optimization, performance, time series prediction

Procedia PDF Downloads 109

Authors: Mohammad Namazi, Mohammad Sadeghzadeh Maharluie

Abstract:

Predicting earnings management is vital for the capital market participants, financial analysts and managers. The aim of this research is attempting to respond to this query: Is there a significant difference between the regression model and neural networks’ models in predicting earnings management, and which one leads to a superior prediction of it? In approaching this question, a Linear Regression (LR) model was compared with two neural networks including Multi-Layer Perceptron (MLP), and Generalized Regression Neural Network (GRNN). The population of this study includes 94 listed companies in Tehran Stock Exchange (TSE) market from 2003 to 2011. After the results of all models were acquired, ANOVA was exerted to test the hypotheses. In general, the summary of statistical results showed that the precision of GRNN did not exhibit a significant difference in comparison with MLP. In addition, the mean square error of the MLP and GRNN showed a significant difference with the multi variable LR model. These findings support the notion of nonlinear behavior of the earnings management. Therefore, it is more appropriate for capital market participants to analyze earnings management based upon neural networks techniques, and not to adopt linear regression models.

Keywords: earnings management, generalized linear regression, neural networks multi-layer perceptron, Tehran stock exchange

Procedia PDF Downloads 421

Authors: Bryan Crompton, Daniel Giger, Tanay Mehta, Apurva Mody

Abstract:

The task of classifying radio frequency (RF) signals has seen recent success in employing deep neural network models. In this work, we present a combined signal processing and machine learning approach to signal classification for cochannel anomalous signals. The power spectral density and cyclostationary signal processing features of a captured signal are computed and fed into a neural net to produce a classification decision. Our combined signal preprocessing and machine learning approach allows for simpler neural networks with fast training times and small computational resource requirements for inference with longer preprocessing time.

Keywords: signal processing, machine learning, cyclostationary signal processing, signal classification

Procedia PDF Downloads 107

Authors: Fathi Abid, Bilel Kaffel

Abstract:

The olive tree, the olive harvest during winter season and the production of olive oil better known by professionals under the name of the crushing operation have interested institutional traders such as olive-oil offices and private companies such as food industry refining and extracting pomace olive oil as well as export-import public and private companies specializing in olive oil. The major problem facing producers of olive oil each winter campaign, contrary to what is expected, it is not whether the harvest will be good or not but whether the sale price will allow them to cover production costs and achieve a reasonable margin of profit or not. These questions are entirely legitimate if we judge by the importance of the issue and the heavy complexity of the uncertainty and competition made tougher by a high level of indebtedness and the experience and expertise of speculators and producers whose objectives are sometimes conflicting. The aim of this paper is to study the formation mechanism of olive oil prices in order to learn about speculators’ behavior and expectations in the market, how they contribute by their industry knowledge and their financial alliances and the size the financial challenge that may be involved for them to build private information hoses globally to take advantage. The methodology used in this paper is based on two stages, in the first stage we study econometrically the formation mechanisms of olive oil price in order to understand the market participant behavior by implementing ARMA, SARMA, GARCH and stochastic diffusion processes models, the second stage is devoted to prediction purposes, we use a combined wavelet- ANN approach. Our main findings indicate that olive oil market participants interact with each other in a way that they promote stylized facts formation. The unstable participant’s behaviors create the volatility clustering, non-linearity dependent and cyclicity phenomena. By imitating each other in some periods of the campaign, different participants contribute to the fat tails observed in the olive oil price distribution. The best prediction model for the olive oil price is based on a back propagation artificial neural network approach with input information based on wavelet decomposition and recent past history.

Keywords: olive oil price, stylized facts, ARMA model, SARMA model, GARCH model, combined wavelet-artificial neural network, continuous-time stochastic volatility mode

Procedia PDF Downloads 339

Authors: Raida Zouari, Iness Ahriz, Rafik Zayani, Ali Dziri, Ridha Bouallegue

Abstract:

This paper proposes a room level indoor localization algorithm based on the use Multi-Layer Neural Network (MLNN) classifiers and one versus one strategy. Seven parameters of the Channel Impulse Response (CIR) were used and Gram-Shmidt Orthogonalization was performed to study the relevance of the extracted parameters. Simulation results show that when relevant CIR parameters are used as position fingerprint and when optimal MLNN architecture is selected good room level localization score can be achieved. The current study showed also that some of the CIR parameters are not correlated to the location and can decrease the localization performance of the system.

Keywords: mobile indoor localization, multi-layer neural network (MLNN), channel impulse response (CIR), Gram-Shmidt orthogonalization

Procedia PDF Downloads 357

Authors: E. A. Mlybari, M. S. Elbisy, A. H. Alshahri, O. M. Albarakati

Abstract:

Sea level rise threatens to increase the impact of future storms and hurricanes on coastal communities. Accurate sea level change prediction and supplement is an important task in determining constructions and human activities in coastal and oceanic areas. In this study, support vector machines (SVM) is proposed to predict daily tidal levels along the Jeddah Coast, Saudi Arabia. The optimal parameter values of kernel function are determined using a genetic algorithm. The SVM results are compared with the field data and with back propagation (BP). Among the models, the SVM is superior to BPNN and has better generalization performance.

Keywords: tides, prediction, support vector machines, genetic algorithm, back-propagation neural network, risk, hazards

Procedia PDF Downloads 468

Authors: Constanza A. Barriga, Rafael Bernardi, Amokrane Berdja, Christian D. Guzman

Abstract:

Most image restoration methods in astronomy rely upon probabilistic tools that infer the best solution for a deconvolution problem. They achieve good performances when the point spread function (PSF) is spatially invariable in the image plane. However, this latter condition is not always satisfied with real optical systems. PSF angular variations cannot be evaluated directly from the observations, neither be corrected at a pixel resolution. We have developed a method for the restoration of images affected by static and anisotropic aberrations using deep neural networks that can be directly applied to sky images. The network is trained using simulated sky images corresponding to the T-80 telescope optical system, an 80 cm survey imager at Cerro Tololo (Chile), which are synthesized using a Zernike polynomial representation of the optical system. Once trained, the network can be used directly on sky images, outputting a corrected version of the image, which has a constant and known PSF across its field-of-view. The method was tested with the T-80 telescope, achieving better results than with PSF deconvolution techniques. We present the method and results on this telescope.

Keywords: aberrations, deep neural networks, image restoration, variable point spread function, wide field images

Procedia PDF Downloads 134

Authors: Yusuf Seedat

Abstract:

The necessity to understand global financial markets has increased following the unfortunate spread of the recent financial crisis around the world. Financial markets are considered to be complex systems consisting of highly volatile move-ments whose indexes fluctuate without any clear pattern. Analytic methods of stock prices have been proposed in which financial markets are modeled using common network analysis tools and methods. It has been found that two key components of social network analysis are relevant to modeling financial markets, allowing us to forecast accurate predictions of stock prices within the financial market. Financial markets have a number of interacting components, leading to complex behavioral patterns. This paper describes a social network approach to analyzing financial markets as a viable approach to studying the way complex stock markets function. We also look at how social network analysis techniques and metrics are used to gauge an understanding of the evolution of financial markets as well as how community detection can be used to qualify and quantify in-fluence within a network.

Keywords: network analysis, social networks, financial markets, stocks, nodes, edges, complex networks

Procedia PDF Downloads 191

Authors: Pranav Gunhal., Krish Jhurani

Abstract:

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

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

Procedia PDF Downloads 118

Authors: H. Y. Jung, Jing Wang, J. H. Park, Hao Shen

Abstract:

This brief addresses the passivity problem for neural networks with time-varying delays. The aim is focus on establishing the passivity condition of the considered neural networks.

Keywords: neural networks, passivity analysis, time-varying delays, linear matrix inequality

Procedia PDF Downloads 570

Authors: Pocholo Rodriguez, Anne Bernadine Ocampo, Ian Benedict Chan, Janric Micah Gray

Abstract:

The state of an electric power system may be classified as either stable or unstable. The borderline of stability is at any condition for which a slight change in an unfavourable direction of any pertinent quantity will cause instability. Voltage instability in power distribution systems could lead to voltage collapse and thus power blackouts. The researchers will present an intelligent system using back propagation algorithm that can detect voltage instability and output voltage of a power distribution and classify it as stable or unstable. The researchers’ work is the use of parameters involved in voltage instability as input parameters to the neural network for training and testing purposes that can provide faster detection and monitoring of the power distribution system.

Keywords: back-propagation algorithm, load instability, neural network, power distribution system

Procedia PDF Downloads 435