Search results for: real time data

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 7

Authors: A. D. Abdellatif, A. N. Ahmed, M. E. Abdelaziz

Abstract:

The main purpose of this paper is to design a double acceptance sampling plan under the time truncated life test when the product lifetime follows an exponentiated transmuted Weibull distribution. Here, the motive is to meet both the consumer’s risk and producer’s risk simultaneously at the specified quality levels, while the termination time is specified. A comparison between the results of the double and single acceptance sampling plans is conducted. We demonstrate the applicability of our results to real data sets.

Keywords: double sampling plan, single sampling plan, producer’s risk, consumer’s risk, exponentiated transmuted weibull distribution, time truncated experiment, single, double, Marshal-Olkin

Procedia PDF Downloads 458

Authors: Nadeem Qaiser, Sherjeel Munsif Khan, Muhammad Mustafa Hussian, Vincent Tung

Abstract:

One of the emerging fields in the detection of chronic diseases is based on the point-of-care (POC) early monitoring of the symptoms and thus provides a state-of-the-art personalized healthcare system. Nowadays, wearable and flexible sensors are being used for analyzing sweat, glucose, blood pressure, and other skin conditions. However, localized jaw-bone swelling called parotid-swelling caused by some viruses has never been tracked before. To track physical motion or deformations, strain sensors, especially piezoresistive ones, are widely used. This work, for the first time, reports carbon nanotubes (CNTs)-based piezoresistive sensing patch that is highly flexible and stretchable and can record muscle deformations in real-time. The developed patch offers an excellent gauge factor for in-plane stretching and spatial expansion with low hysteresis. To calibrate the volumetric muscle expansion, we fabricated the pneumatic actuator that experienced volumetric expansion and thus redefined the gauge factor. Moreover, we employ a Bluetooth-low-energy system that can send information about muscle activity in real-time to a smartphone app. We utilized COMSOL calculations to reveal the mechanical robustness of the patch. The experiments showed the sensing patch's greater cyclability, making it a patch for personal healthcare and an excellent choice for monitoring the real-time POC monitoring of the human muscle swelling.

Keywords: piezoresistive strain sensor, FEM simulations, CNTs sensor, flexible

Procedia PDF Downloads 63

Authors: Tokihiko Akita, Seiichi Mita

Abstract:

A millimeter-wave radar is the most robust against adverse environments, making it an essential environment recognition sensor for automated driving. However, the reflection signal is sparse and unstable, so it is difficult to obtain the high recognition accuracy. Deep learning provides high accuracy even for them in recognition, but requires large scale datasets with ground truth. Specially, it takes a lot of cost to annotate for a millimeter-wave radar. For the solution, utilizing a simulator that can generate an annotated huge dataset is effective. Simulation of the radar is more difficult to match with real world data than camera image, and recognition by deep learning with higher-order features using the simulator causes further deviation. We have challenged to improve the accuracy of traffic participant classification by fusing simulator and real-world data with domain adaptation technique. Experimental results with the domain adaptation network created by us show that classification accuracy can be improved even with a few real-world data.

Keywords: millimeter-wave radar, object classification, deep learning, simulation, domain adaptation

Procedia PDF Downloads 66

Authors: Zhdanova A. O., Volkov R. S., Kuznetsov G. V., Strizhak P. A.

Abstract:

Fires in different types of facilities are a serious problem worldwide.It is still an unaccomplished science and technology objective to establish the minimum number and type of sensors in automatic systems of compartment fire suppression which would turn the fire-extinguishing agent spraying on and off in real time depending on the state of the fire, minimize the amount of agent applied, delay time in fire suppression and system response, as well as the time of combustion suppression. Based on the results of experimental studies, the conclusion was made that it is reasonable to use a gas analysis system and heat sensors (in the event of their prior activation) to determine the effectiveness of fire suppression (fire-extinguishing composition interacts with the fire). Thus, the concentration of CO in the interaction of the firefighting liquid with the fire increases to 0.7–1.2%, which indicates a slowdown in the flame combustion, and heat sensors stop responding at a gas medium temperature below 80 ºC, which shows a gradual decrease in the heat release from the fire. The evidence from this study suggests that the information received from the video recording equipment (video camera) should be used in real time as an additional parameter confirming fire suppression. Research was supported by Russian Science Foundation (project No 21-19-00009, https://rscf.ru/en/project/21-19-00009/).

Keywords: compartment fires, fire suppression, continuous control of fire behavior, feedback systems

Procedia PDF Downloads 106

Authors: Gerhard Gritsch, Ana Skupch, Manfred Hartmann, Wolfgang Frühwirt, Hannes Perko, Dieter Grossegger, Tilmann Kluge

Abstract:

Rehabilitation of stroke patients is dominated by classical physiotherapy. Nowadays, a field of research is the application of neurofeedback techniques in order to help stroke patients to get rid of their motor impairments. Especially, if a certain limb is completely paralyzed, neurofeedback is often the last option to cure the patient. Certain exercises, like the imagination of the impaired motor function, have to be performed to stimulate the neuroplasticity of the brain, such that in the neighboring parts of the injured cortex the corresponding activity takes place. During the exercises, it is very important to keep the motivation of the patient at a high level. For this reason, the missing natural feedback due to a movement of the effected limb may be replaced by a synthetic feedback based on the motor-related brain function. To generate such a synthetic feedback a system is needed which measures, detects, localizes and visualizes the motor related µ-rhythm. Fast therapeutic success can only be achieved if the feedback features high specificity, comes in real-time and without large delay. We describe such an approach that offers a 3D visualization of µ-rhythms in real time with a delay of 500ms. This is accomplished by combining smart EEG preprocessing in the frequency domain with source localization techniques. The algorithm first selects the EEG channel featuring the most prominent rhythm in the alpha frequency band from a so-called motor channel set (C4, CZ, C3; CP6, CP4, CP2, CP1, CP3, CP5). If the amplitude in the alpha frequency band of this certain electrode exceeds a threshold, a µ-rhythm is detected. To prevent detection of a mixture of posterior alpha activity and µ-activity, the amplitudes in the alpha band outside the motor channel set are not allowed to be in the same range as the main channel. The EEG signal of the main channel is used as template for calculating the spatial distribution of the µ - rhythm over all electrodes. This spatial distribution is the input for a inverse method which provides the 3D distribution of the µ - activity within the brain which is visualized in 3D as color coded activity map. This approach mitigates the influence of lid artifacts on the localization performance. The first results of several healthy subjects show that the system is capable of detecting and localizing the rarely appearing µ-rhythm. In most cases the results match with findings from visual EEG analysis. Frequent eye-lid artifacts have no influence on the system performance. Furthermore, the system will be able to run in real-time. Due to the design of the frequency transformation the processing delay is 500ms. First results are promising and we plan to extend the test data set to further evaluate the performance of the system. The relevance of the system with respect to the therapy of stroke patients has to be shown in studies with real patients after CE certification of the system. This work was performed within the project ‘LiveSolo’ funded by the Austrian Research Promotion Agency (FFG) (project number: 853263).

Keywords: real-time EEG neuroimaging, neurofeedback, stroke, EEG–signal processing, rehabilitation

Procedia PDF Downloads 362

Authors: Linda Dusman, Linda Baker

Abstract:

The Bearman Study of Audience Engagement examined the potential for real-time music education during online symphony orchestra concerts. It follows on the promising results of a preliminary study of STEAM (Science, Technology, Engineering, Arts, and Mathematics) education during live concerts, funded by the National Science Foundation with the Baltimore Symphony Orchestra. For the Bearman Study, audience groups were recruited to attend two previously recorded concerts of the National Orchestral Institute (NOI) in 2020 or the Utah Symphony in 2021. They used a smartphone app called EnCue to present real-time program notes about the music being performed. Short notes along with visual information (photos and score fragments) were designed to provide historical, cultural, biographical, and theoretical information at specific moments in the music where that information would be most pertinent, generally spaced 2-3 minutes apart to avoid distraction. The music performed included Dvorak Symphony No. 8 and Mahler Symphony No. 5 at NOI, and Mendelssohn Scottish Symphony and Richard Strauss Metamorphosen with the Utah Symphony, all standard repertoire for symphony orchestras. During each phase of the study (2020 and 2021), participants were randomly assigned to use the app to view program notes during the first concert or to use the app during the second concert. A total of 139 participants (67 in 2020 and 72 in 2021) completed three online questionnaires, one before attending the first concert, one immediately after the concert, and the third immediately after the second concert. Questionnaires assessed demographic background, expertise in music, engagement during the concert, learning of content about the composers and the symphonies, and interest in the future use of the app. In both phases of the study, participants demonstrated that they learned content presented on the app, evidenced by the fact that their multiple-choice test scores were significantly higher when they used the app than when they did not. In addition, most participants indicated that using the app enriched their experience of the concert. Overall, they were very positive about their experience using the app for real-time learning and they expressed interest in using it in the future at both live and streaming concerts. Results confirmed that informal real-time learning during concerts is possible and can generate enhanced engagement and interest in classical music.

Keywords: audience engagement, informal education, music technology, real-time learning

Procedia PDF Downloads 179

Authors: Duy Nhat Tran, Van Tien Pham, Quang Trung Trinh, Tien Hai Tran, Van Cong Bui

Abstract:

In recent years, artificial intelligence and the Internet of Things have experienced significant advancements. Collecting image data and real-time analysis and processing of tasks have become increasingly popular in various aspects of life. Optical scanning devices are widely used to observe and analyze different environments, whether fixed outdoors, mounted on mobile devices, or used in unmanned aerial vehicles. As a result, the interaction between the physical environment and these devices has become more critical in terms of safety. Two commonly used methods for addressing these challenges are active and passive approaches. Each method has its advantages and disadvantages, but combining both methods can lead to higher efficiency. One solution is to utilize direct-drive motors for position control and real-time feedback within the operational range to determine appropriate control parameters with high precision. If the maximum motor torque is smaller than the inertial torque and the rotor reaches the operational limit, the spring system absorbs the impact force. Numerous experiments have been conducted to demonstrate the effectiveness of device protection during operation.

Keywords: optical device, collision safety, collision absorption, precise mechanics

Procedia PDF Downloads 34

Authors: Emilienne Lardy, Mariam Lafkihi, Eric Ballot

Abstract:

Efficient city logistics requires a comprehensive understanding of traffic dynamics, particularly as it pertains to kinetic parameters influencing energy consumption and trip duration estimations. While real-time traffic information is increasingly accessible, current high-precision forecasting services embedded in route planning often function as opaque 'black boxes' for users. These services, typically relying on AI-processed counting data, fall short in accommodating open design parameters essential for management studies, notably within Supply Chain Management. This work revisits the modeling of traffic conditions in the context of city logistics, emphasizing its significance from the user’s point of view, with two focuses. Firstly, the focus is not on the vehicle flow but on the vehicles themselves and the impact of the traffic conditions on their driving behaviour. This means opening the range of studied indicators beyond vehicle speed to describe extensively the kinetic and dynamic aspects of driving behaviour. To achieve this, we leverage the Art. Kinema parameters are designed to characterize driving cycles. Secondly, this study examines how the driving context (i.e., exogenous factors to the traffic flow) determines the mentioned driving behaviour. Specifically, we explore it investigates how accurately the kinetic behaviour of a vehicle can be predicted based on a limited set of exogenous factors, such as time, day, road type, orientation, slope, and weather conditions. To answer this question, statistical analysis was conducted on real-world driving data, which includes high-frequency measurements of vehicle speed. A Generalized Linear Model has been established to link kinetic parameters with independent categorical contextual variables. The results include the analysis of the regression’s accuracy, as well as the utilization of the regression’s results in freight distribution scenario elaboration and analysis for Supply Chain Management purposes.

Keywords: driving context, generalized linear model, kinetic behaviour, real world driving data

Procedia PDF Downloads 40

Authors: Moung Young Lee, Chul Gyu Song

Abstract:

Photoacoustic imaging is the imaging technology that combines the optical imaging and ultrasound. This provides the high contrast and resolution due to optical imaging and ultrasound imaging, respectively. We developed the real-time photoacoustic tomography (PAT) system using linear-ultrasound transducer and digital acquisition (DAQ) board. There are two types of algorithm for reconstructing the photoacoustic signal. One is back-projection algorithm, the other is FFT algorithm. Especially, we used the non-uniform FFT algorithm. To evaluate the performance of our system and algorithms, we monitored two wires that stands at interval of 2.89 mm and 0.87 mm. Then, we compared the images reconstructed by algorithms. Finally, we monitored the two hairs crossed and compared between these algorithms.

Keywords: back-projection, image comparison, non-uniform FFT, photoacoustic tomography

Procedia PDF Downloads 411

Authors: The Danh Phan

Abstract:

House price forecasting is a main topic in the real estate market research. Effective house price prediction models could not only allow home buyers and real estate agents to make better data-driven decisions but may also be beneficial for the property policymaking process. This study investigates the housing market by using machine learning techniques to analyze real historical house sale transactions in Australia. It seeks useful models which could be deployed as an application for house buyers and sellers. Data analytics show a high discrepancy between the house price in the most expensive suburbs and the most affordable suburbs in the city of Melbourne. In addition, experiments demonstrate that the combination of Stepwise and Support Vector Machine (SVM), based on the Mean Squared Error (MSE) measurement, consistently outperforms other models in terms of prediction accuracy.

Keywords: house price prediction, regression trees, neural network, support vector machine, stepwise

Procedia PDF Downloads 194

Authors: J. Ierache, N. A. Mangiarua, S. A. Bevacqua, N. N. Verdicchio, M. E. Becerra, D. R. Sanz, M. E. Sena, F. M. Ortiz, N. D. Duarte, S. Igarza

Abstract:

Augmented Reality is a technology that involves the overlay of virtual content, which is context or environment sensitive, on images of the physical world in real time. This paper presents the development of a catalog system that facilitates and allows the creation, publishing, management and exploitation of augmented multimedia contents and Augmented Reality applications, creating an own space for anyone that wants to provide information to real objects in order to edit and share it then online with others. These spaces would be built for different domains without the initial need of expert users. Its operation focuses on the context of Web 2.0 or Social Web, with its various applications, developing contents to enrich the real context in which human beings act permitting the evolution of catalog’s contents in an emerging way.

Keywords: augmented reality, catalog system, computer graphics, mobile application

Procedia PDF Downloads 330

Authors: Semanur Savruk, Onur Akbatı

Abstract:

Integration of various communication protocols into an automation system causes a rise in setup and maintenance cost and make to control network devices in difficulty. The gateway becomes necessary for reducing complexity in network topology. In this study, Modbus RTU/Modbus TCP industrial ethernet gateway design and implementation are presented with ARM embedded system and FreeRTOS real-time operating system. The Modbus gateway can perform communication with Modbus RTU and Modbus TCP devices over itself. Moreover, the gateway can be adjustable with the user-interface application or messaging interface. Conducted experiments and the results are presented in the paper. Eventually, the proposed system is a complete, low-cost, real-time, and user-friendly design for monitoring and setting devices and useful for meeting remote control purposes.

Keywords: gateway, industrial communication, modbus, network

Procedia PDF Downloads 115

Authors: Qiang Zhang, Xiaojian Hu

Abstract:

In multi-vehicle tracking, based on feature selection, the tracking system efficiently tracks vehicles in a video with minimal error in combination with color attribution, which focuses on presenting a simple and fast, yet accurate and robust solution to the problem such as inaccurately and untimely responses of statistics-based adaptive traffic control system in the intersection scenario. In this study, a real-time tracking system is proposed for multi-vehicle tracking in the intersection scene. Considering the complexity and application feasibility of the algorithm, in the object detection step, the detection result provided by virtual loops were post-processed and then used as the input for the tracker. For the tracker, lightweight methods were designed to extract and select features and incorporate them into the adaptive color tracking (ACT) framework. And the approbatory online feature selection algorithms are integrated on the mature ACT system with good compatibility. The proposed feature selection methods and multi-vehicle tracking method are evaluated on KITTI datasets and show efficient vehicle tracking performance when compared to the other state-of-the-art approaches in the same category. And the system performs excellently on the video sequences recorded at the intersection. Furthermore, the presented vehicle tracking system is suitable for surveillance applications.

Keywords: real-time, multi-vehicle tracking, feature selection, color attribution

Procedia PDF Downloads 132

Authors: Mohamed Mounir Tellache, Hiroyuki Kambara, Yasuharu Koike, Makoto Miyakoshi, Natsue Yoshimura

Abstract:

This study explores the practicality of using electroencephalographic (EEG) independent components to predict eight-direction finger movements in pseudo-real-time. Six healthy participants with individual-head MRI images performed finger movements in eight directions with two different arm configurations. The analysis was performed in two stages. The first stage consisted of using independent component analysis (ICA) to separate the signals representing brain activity from non-brain activity signals and to obtain the unmixing matrix. The resulting independent components (ICs) were checked, and those reflecting brain-activity were selected. Finally, the time series of the selected ICs were used to predict eight finger-movement directions using Sparse Logistic Regression (SLR). The second stage consisted of using the previously obtained unmixing matrix, the selected ICs, and the model obtained by applying SLR to classify a different EEG dataset. This method was applied to two different settings, namely the single-participant level and the group-level. For the single-participant level, the EEG dataset used in the first stage and the EEG dataset used in the second stage originated from the same participant. For the group-level, the EEG datasets used in the first stage were constructed by temporally concatenating each combination without repetition of the EEG datasets of five participants out of six, whereas the EEG dataset used in the second stage originated from the remaining participants. The average test classification results across datasets (mean ± S.D.) were 38.62 ± 8.36% for the single-participant, which was significantly higher than the chance level (12.50 ± 0.01%), and 27.26 ± 4.39% for the group-level which was also significantly higher than the chance level (12.49% ± 0.01%). The classification accuracy within [–45°, 45°] of the true direction is 70.03 ± 8.14% for single-participant and 62.63 ± 6.07% for group-level which may be promising for some real-life applications. Clustering and contribution analyses further revealed the brain regions involved in finger movement and the temporal aspect of their contribution to the classification. These results showed the possibility of using the ICA-based method in combination with other methods to build a real-time system to control prostheses.

Keywords: brain-computer interface, electroencephalography, finger motion decoding, independent component analysis, pseudo real-time motion decoding

Procedia PDF Downloads 119

Authors: Alejandro Rodríguez-Molinero, Carlos Pérez-López, Jorge Hernández-Vara, Àngels Bayes-Rusiñol, Juan Carlos Martínez-Castrillo, David A. Pérez-Martínez

Abstract:

Background: Monitoring motor symptoms in Parkinson's patients is often a complex and time-consuming task for clinicians, as Hauser's diaries are often poorly completed by patients. Recently, new automatic devices (Parkinson's holter: STAT-ON®) have been developed capable of monitoring patients' motor fluctuations. The MoMoPa-EC clinical trial (NCT04176302) investigates which of the two methods produces better clinical results. In this sub-analysis, the concordance between both methods is analyzed. Methods: In the MoMoPa-EC clinical trial, 164 patients with moderate-severe Parkinson's disease and at least two hours a day of Off will be included. At the time of patient recruitment, all of them completed a seven-day motor fluctuation diary at home (Hauser’s diary) while wearing the Parkinson's holter. In this sub-analysis, 71 patients with complete data for the purpose of this comparison were included. The intraclass correlation coefficient was calculated between the patient diary entries and the Parkinson's holter data in terms of time On, Off, and time with dyskinesias. Results: The intra-class correlation coefficient of both methods was 0.57 (95% CI: 0.3-0.74) for daily time in Off (%), 0.48 (95% CI: 0.14-0.68) for daily time in On (%), and 0.37 (95% CI %: -0.04-0.62) for daily time with dyskinesias (%). Conclusions: Both methods have a moderate agreement with each other. We will have to wait for the results of the MoMoPa-EC project to estimate which of them has the greatest clinical benefits. Acknowledgment: This work is supported by AbbVie S.L.U, the Instituto de Salud Carlos III [DTS17/00195], and the European Fund for Regional Development, 'A way to make Europe'.

Keywords: Parkinson, sensor, motor fluctuations, dyskinesia

Procedia PDF Downloads 204

Authors: Khalid Bashir, Hammad Zahid

Abstract:

Most of the investors consider that economic and financial information is the most important at the time of making investment decisions. But it is not true, as in the past two decades, the Behavioral aspects and the behavioral biases have gained an important place in the decision-making process of an investor. This study is basically conducted on this fact. The purpose of this study is to examine the impact of behavioral factors on the decision-making of the individual real estate investor in Pakistan. Some important behavioral factors like overconfidence, anchoring, gambler’s fallacy, home bias, loss aversion, regret aversion, mental accounting, herding and representativeness are used in this study to find their impact on the psychology of individual investors. The targeted population is the real estate investor of Pakistan, and a sample of 650 investors is selected on the basis of convenience sampling technique. The data is collected through the questionnaire with a response rate of 46.15 %. Descriptive statistical techniques and SEM are used to analyze the data by using statistical software. The results revealed the fact that some behavioral factors have a significant impact on the decision-making of investors. Among all the behavioral biases, overconfidence, anchoring, gambler’s fallacy, loss aversion and representativeness have a significant positive impact on the decision-making of the individual investor, while the rest of biases like home bias, regret aversion, mental accounting, herding have less impact on the decision-making process of an individual.

Keywords: behavioral finance, anchoring, gambler’s fallacy, loss aversion

Procedia PDF Downloads 40

Authors: Lindelwa Mngomezulu, Tonderai Muchenje

Abstract:

Over the decades, E-mails provide easy, fast and timely communication enabling businesses and state owned agencies to communicate with their stakeholders and with their own employees in real-time. Moreover, since the launch of Microsoft office 365 and many other clouds based E-mail services, many businesses have been migrating from the on premises E-mail services to the cloud and more precisely since the beginning of the Covid-19 pandemic, there has been a significant increase of E-mails utilization, which then leads to the increase of cyber-attacks. In that regard, E-mail security has become very important in the E-mail transportation to ensure that the E-mail gets to the recipient without the data integrity being compromised. The classification of the features to enhance E-mail security for further from the enhanced cyber-attacks as we are aware that since the technology is advancing so at the cyber-attacks. Therefore, in order to maximize the data integrity we need to also maximize security of the E-mails such as enhanced E-mail authentication. The successful enhancement of E-mail security in the future may lessen the frequency of information thefts via E-mails, resulting in the data of South African State-owned agencies not being compromised.

Keywords: e-mail security, cyber-attacks, data integrity, authentication

Procedia PDF Downloads 107

Authors: Gayathri Nagarajan, L. D. Dhinesh Babu

Abstract:

Health care is one of the prominent industries that generate voluminous data thereby finding the need of machine learning techniques with big data solutions for efficient processing and prediction. Missing data, incomplete data, real time streaming data, sensitive data, privacy, heterogeneity are few of the common challenges to be addressed for efficient processing and mining of health care data. In comparison with other applications, accuracy and fast processing are of higher importance for health care applications as they are related to the human life directly. Though there are many machine learning techniques and big data solutions used for efficient processing and prediction in health care data, different techniques and different frameworks are proved to be effective for different applications largely depending on the characteristics of the datasets. In this paper, we present a framework that uses ensemble machine learning technique gradient boosted trees for data classification in health care big data. The framework is built on Spark platform which is fast in comparison with other traditional frameworks. Unlike other works that focus on a single technique, our work presents a comparison of six different machine learning techniques along with gradient boosted trees on datasets of different characteristics. Five benchmark health care datasets are considered for experimentation, and the results of different machine learning techniques are discussed in comparison with gradient boosted trees. The metric chosen for comparison is misclassification error rate and the run time of the algorithms. The goal of this paper is to i) Compare the performance of gradient boosted trees with other machine learning techniques in Spark platform specifically for health care big data and ii) Discuss the results from the experiments conducted on datasets of different characteristics thereby drawing inference and conclusion. The experimental results show that the accuracy is largely dependent on the characteristics of the datasets for other machine learning techniques whereas gradient boosting trees yields reasonably stable results in terms of accuracy without largely depending on the dataset characteristics.

Keywords: big data analytics, ensemble machine learning, gradient boosted trees, Spark platform

Procedia PDF Downloads 219

Authors: Khalida Douibi, Rodrigo Balp, Solène Le Bars

Abstract:

In the medical field, applications related to human experiments are frequently linked to reduced samples size, which makes the training of machine learning models quite sensitive and therefore not very robust nor generalizable. This is notably the case in Brain-Computer Interface (BCI) studies, where the sample size rarely exceeds 20 subjects or a few number of trials. To address this problem, several resampling approaches are often used during the data preparation phase, which is an overly critical step in a data science analysis process. One of the naive approaches that is usually applied by data scientists consists in the transformation of the entire database before the resampling phase. However, this can cause model’ s performance to be incorrectly estimated when making predictions on unseen data. In this paper, we explored the effect of data leakage observed during our BCI experiments for device control through the real-time classification of SSVEPs (Steady State Visually Evoked Potentials). We also studied potential ways to ensure optimal validation of the classifiers during the calibration phase to avoid overfitting. The results show that the scaling step is crucial for some algorithms, and it should be applied after the resampling phase to avoid data leackage and improve results.

Keywords: data leackage, data science, machine learning, SSVEP, BCI, overfitting

Procedia PDF Downloads 128

Authors: Andrey Matyukhin

Abstract:

Russian official statistics have been showing a steady decline in residential real estate prices for several consecutive years. Price risk in real estate markets is thus affecting various groups of economic agents, namely, individuals, construction companies and financial institutions. Potential use of property derivatives might help mitigate adverse consequences of negative price dynamics. Unless a sustainable price indicator is developed, settlement of such instruments imposes constraints on counterparties involved while imposing restrictions on real estate market development. The study addresses geographical and classification heterogeneity in real estate prices by means of variance analysis in various groups of real estate properties. In conclusion, we determine optimal sample structure of representative real estate assets with sufficient level of price homogeneity. The composite price indicator based on the sample would have a higher level of robustness and reliability and hence improving liquidity in the market for property derivatives through underlying standardization. Unlike the majority of existing real estate price indices, calculated on country-wide basis, the optimal indices for Russian market shall be constructed on the city-level.

Keywords: price homogeneity, property derivatives, real estate price index, real estate price risk

Procedia PDF Downloads 283

Authors: Rahul Paul, Peter Mctaggart, Luke Skinner

Abstract:

Maintaining public safety and network reliability are the core objectives of all electricity distributors globally. For many electricity distributors, managing vegetation clearances from their above ground assets (poles and conductors) is the most important and costly risk mitigation control employed to meet these objectives. Light Detection And Ranging (LiDAR) is widely used by utilities as a cost-effective method to inspect their spatially-distributed assets at scale, often captured using high powered LiDAR scanners attached to fixed wing or rotary aircraft. The resulting 3D point cloud model is used by these utilities to perform engineering grade measurements that guide the prioritisation of vegetation cutting programs. Advances in computer vision and machine-learning approaches are increasingly applied to increase automation and reduce inspection costs and time; however, real-world LiDAR capture variables (e.g., aircraft speed and height) create complexity, noise, and missing data, reducing the effectiveness of these approaches. This paper proposes a method for identifying each conductor from LiDAR data via clustering methods that can precisely reconstruct conductors in complex real-world configurations in the presence of high levels of noise. It proposes 3D catenary models for individual clusters fitted to the captured LiDAR data points using a least square method. An iterative learning process is used to identify potential conductor models between pole pairs. The proposed method identifies the optimum parameters of the catenary function and then fits the LiDAR points to reconstruct the conductors.

Keywords: point cloud, LİDAR data, machine learning, computer vision, catenary curve, vegetation management, utility industry

Procedia PDF Downloads 66

Authors: Sahraoui Halima, Dahani Ameur, Tigrine Abedelkader

Abstract:

DigiMesh technology represents a pioneering advancement in wireless networking, offering cost-effective and energy-efficient capabilities. Its inherent simplicity and adaptability facilitate the seamless transfer of data between network nodes, extending the range and ensuring robust connectivity through autonomous self-healing mechanisms. In light of these advantages, this study introduces a medical platform harnessed with DigiMesh wireless network technology characterized by low power consumption, immunity to interference, and user-friendly operation. The primary application of this platform is the real-time, long-distance monitoring of Electrocardiogram (ECG) signals, with the added capacity for simultaneous monitoring of ECG signals from multiple patients. The experimental setup comprises key components such as Raspberry Pi, E-Health Sensor Shield, and Xbee DigiMesh modules. The platform is composed of multiple ECG acquisition devices labeled as Sensor Node 1 and Sensor Node 2, with a Raspberry Pi serving as the central hub (Sink Node). Two communication approaches are proposed: Single-hop and multi-hop. In the Single-hop approach, ECG signals are directly transmitted from a sensor node to the sink node through the XBee3 DigiMesh RF Module, establishing peer-to-peer connections. This approach was tested in the first experiment to assess the feasibility of deploying wireless sensor networks (WSN). In the multi-hop approach, two sensor nodes communicate with the server (Sink Node) in a star configuration. This setup was tested in the second experiment. The primary objective of this research is to evaluate the performance of both Single-hop and multi-hop approaches in diverse scenarios, including open areas and obstructed environments. Experimental results indicate the DigiMesh network's effectiveness in Single-hop mode, with reliable communication over distances of approximately 300 meters in open areas. In the multi-hop configuration, the network demonstrated robust performance across approximately three floors, even in the presence of obstacles, without the need for additional router devices. This study offers valuable insights into the capabilities of DigiMesh wireless technology for real-time ECG monitoring in healthcare applications, demonstrating its potential for use in diverse medical scenarios.

Keywords: DigiMesh protocol, ECG signal, real-time monitoring, medical platform

Procedia PDF Downloads 51

Authors: Merve Uzuner Sahin, Kumru D. Atalay, Berna Dengiz

Abstract:

In today’s competitive market, there are many alternative products that can be used in similar manner and purpose. Therefore, the utility of the product is an important issue for the preferability of the brand. This utility could be measured in terms of its functionality, durability, reliability. These all are affected by the system capabilities. Reliability is an important system design criteria for the manufacturers to be able to have high availability. Availability is the probability that a system (or a component) is operating properly to its function at a specific point in time or a specific period of times. System availability provides valuable input to estimate the production rate for the company to realize the production plan. When considering only the corrective maintenance downtime of the system, mean time between failure (MTBF) and mean time to repair (MTTR) are used to obtain system availability. Also, the MTBF and MTTR values are important measures to improve system performance by adopting suitable maintenance strategies for reliability engineers and practitioners working in a system. Failure and repair time probability distributions of each component in the system should be known for the conventional availability analysis. However, generally, companies do not have statistics or quality control departments to store such a large amount of data. Real events or situations are defined deterministically instead of using stochastic data for the complete description of real systems. A fuzzy set is an alternative theory which is used to analyze the uncertainty and vagueness in real systems. The aim of this study is to present a novel approach to compute system availability using representation of MTBF and MTTR in fuzzy numbers. Based on the experience in the system, it is decided to choose 3 different spread of MTBF and MTTR such as 15%, 20% and 25% to obtain lower and upper limits of the fuzzy numbers. To the best of our knowledge, the proposed method is the first application that is used fuzzy MTBF and fuzzy MTTR for fuzzy system availability estimation. This method is easy to apply in any repairable production system by practitioners working in industry. It is provided that the reliability engineers/managers/practitioners could analyze the system performance in a more consistent and logical manner based on fuzzy availability. This paper presents a real case study of a repairable multi-stage production line in lead-acid battery production factory in Turkey. The following is focusing on the considered wet-charging battery process which has a higher production level than the other types of battery. In this system, system components could exist only in two states, working or failed, and it is assumed that when a component in the system fails, it becomes as good as new after repair. Instead of classical methods, using fuzzy set theory and obtaining intervals for these measures would be very useful for system managers, practitioners to analyze system qualifications to find better results for their working conditions. Thus, much more detailed information about system characteristics is obtained.

Keywords: availability analysis, battery production system, fuzzy sets, triangular fuzzy numbers (TFNs)

Procedia PDF Downloads 196

Authors: Wei Li, Tuo Yang

Abstract:

In order to satisfy the real-time and accurate requirements of cigarette smoke detection in complex scenes, a cigarette smoke detection technology based on the combination of deep learning and color features was proposed. Firstly, based on the color features of cigarette smoke, the suspicious cigarette smoke area in the image is extracted. Secondly, combined with the efficiency of cigarette smoke detection and the problem of network overfitting, a network model for cigarette smoke detection was designed according to YOLOV3 algorithm to reduce the false detection rate. The experimental results show that the method is feasible and effective, and the accuracy of cigarette smoke detection is up to 99.13%, which satisfies the requirements of real-time cigarette smoke detection in complex scenes.

Keywords: deep learning, computer vision, cigarette smoke detection, YOLOV3, color feature extraction

Procedia PDF Downloads 60

Authors: Parampal Kaur, Deepak Aggarwal

Abstract:

In the computational Grid, fault tolerance is an imperative issue to be considered during job scheduling. Due to the widespread use of resources, systems are highly prone to errors and failures. Hence, fault tolerance plays a key role in the grid to avoid the problem of unreliability. Scheduling the task to the appropriate resource is a vital requirement in computational Grid. The fittest resource scheduling algorithm searches for the appropriate resource based on the job requirements, in contrary to the general scheduling algorithms where jobs are scheduled to the resources with best performance factor. The proposed method is to improve the fault tolerance of the fittest resource scheduling algorithm by scheduling the job in coordination with job replication when the resource has low reliability. Based on the reliability index of the resource, the resource is identified as critical. The tasks are scheduled based on the criticality of the resources. Results show that the execution time of the tasks is comparatively reduced with the proposed algorithm using real-time approach rather than a simulator.

Keywords: computational grid, fault tolerance, task replication, job scheduling

Procedia PDF Downloads 412

Authors: Ali Kazemi

Abstract:

The digital evolution of financial services, while offering unprecedented convenience and accessibility, has also escalated the vulnerabilities to fraudulent activities. In this study, we introduce a distinct approach to real-time anomaly detection in financial transactions, aiming to fortify the defenses of banking and financial institutions against such threats. Utilizing unsupervised machine learning algorithms, specifically autoencoders and isolation forests, our research focuses on identifying irregular patterns indicative of fraud within transactional data, thus enabling immediate action to prevent financial loss. The data we used in this study included the monetary value of each transaction. This is a crucial feature as fraudulent transactions may have distributions of different amounts than legitimate ones, such as timestamps indicating when transactions occurred. Analyzing transactions' temporal patterns can reveal anomalies (e.g., unusual activity in the middle of the night). Also, the sector or category of the merchant where the transaction occurred, such as retail, groceries, online services, etc. Specific categories may be more prone to fraud. Moreover, the type of payment used (e.g., credit, debit, online payment systems). Different payment methods have varying risk levels associated with fraud. This dataset, anonymized to ensure privacy, reflects a wide array of transactions typical of a global banking institution, ranging from small-scale retail purchases to large wire transfers, embodying the diverse nature of potentially fraudulent activities. By engineering features that capture the essence of transactions, including normalized amounts and encoded categorical variables, we tailor our data to enhance model sensitivity to anomalies. The autoencoder model leverages its reconstruction error mechanism to flag transactions that deviate significantly from the learned normal pattern, while the isolation forest identifies anomalies based on their susceptibility to isolation from the dataset's majority. Our experimental results, validated through techniques such as k-fold cross-validation, are evaluated using precision, recall, and the F1 score alongside the area under the receiver operating characteristic (ROC) curve. Our models achieved an F1 score of 0.85 and a ROC AUC of 0.93, indicating high accuracy in detecting fraudulent transactions without excessive false positives. This study contributes to the academic discourse on financial fraud detection and provides a practical framework for banking institutions seeking to implement real-time anomaly detection systems. By demonstrating the effectiveness of unsupervised learning techniques in a real-world context, our research offers a pathway to significantly reduce the incidence of financial fraud, thereby enhancing the security and trustworthiness of digital financial services.

Keywords: anomaly detection, financial fraud, machine learning, autoencoders, isolation forest, transactional data analysis

Procedia PDF Downloads 26

Authors: Faizul Azly Abd Dzubir, Muhammad F. Othman

Abstract:

Conventional structural health monitoring approach for mechanical equipment uses inspection data from Non-Destructive Testing (NDT) during plant shut down window and fitness for service evaluation to estimate the integrity of the equipment that is prone to crack damage. Yet, this forecast is fraught with uncertainty because it is often based on assumptions of future operational parameters, and the prediction is not continuous or online. Advanced Diagnostic and Prognostic Technology (ADaPT) uses Acoustic Emission (AE) technology and a stochastic prognostic model to provide real-time monitoring and prediction of mechanical defects or cracks. The forecast can help the plant authority handle their cracked equipment before it ruptures, causing an unscheduled shutdown of the facility. The ADaPT employs process historical data trending, finite element analysis, fitness for service, and probabilistic statistical analysis to develop a prediction model for crack initiation and growth due to mechanical damage. The prediction model is combined with live equipment operating data for real-time prediction of the remaining life span owing to fracture. ADaPT was devised at a hot combined feed exchanger (HCFE) that had suffered creep crack damage. The ADaPT tool predicts the initiation of a crack at the top weldment area by April 2019. During the shutdown window in April 2019, a crack was discovered and repaired. Furthermore, ADaPT successfully advised the plant owner to run at full capacity and improve output by up to 7% by April 2019. ADaPT was also used on a coke drum that had extensive fatigue cracking. The initial cracks are declared safe with ADaPT, with remaining crack lifetimes extended another five (5) months, just in time for another planned facility downtime to execute repair. The prediction model, when combined with plant information data, allows plant operators to continuously monitor crack propagation caused by mechanical damage for improved maintenance planning and to avoid costly shutdowns to repair immediately.

Keywords: mechanical damage, cracks, continuous monitoring tool, remaining life, acoustic emission, prognostic model

Procedia PDF Downloads 52

Authors: Stefan K. Behfar

Abstract:

The rapid growth of the Ethereum network has brought forth the urgent need for scalable analysis methods to handle the increasing volume of blockchain data. In this research, we propose efficient methodologies for making Ethereum network analysis scalable. Our approach leverages a combination of graph-based data representation, probabilistic sampling, and parallel processing techniques to achieve unprecedented scalability while preserving critical network insights. Data Representation: We develop a graph-based data representation that captures the underlying structure of the Ethereum network. Each block transaction is represented as a node in the graph, while the edges signify temporal relationships. This representation ensures efficient querying and traversal of the blockchain data. Probabilistic Sampling: To cope with the vastness of the Ethereum blockchain, we introduce a probabilistic sampling technique. This method strategically selects a representative subset of transactions and blocks, allowing for concise yet statistically significant analysis. The sampling approach maintains the integrity of the network properties while significantly reducing the computational burden. Graph Convolutional Networks (GCNs): We incorporate GCNs to process the graph-based data representation efficiently. The GCN architecture enables the extraction of complex spatial and temporal patterns from the sampled data. This combination of graph representation and GCNs facilitates parallel processing and scalable analysis. Distributed Computing: To further enhance scalability, we adopt distributed computing frameworks such as Apache Hadoop and Apache Spark. By distributing computation across multiple nodes, we achieve a significant reduction in processing time and enhanced memory utilization. Our methodology harnesses the power of parallelism, making it well-suited for large-scale Ethereum network analysis. Evaluation and Results: We extensively evaluate our methodology on real-world Ethereum datasets covering diverse time periods and transaction volumes. The results demonstrate its superior scalability, outperforming traditional analysis methods. Our approach successfully handles the ever-growing Ethereum data, empowering researchers and developers with actionable insights from the blockchain. Case Studies: We apply our methodology to real-world Ethereum use cases, including detecting transaction patterns, analyzing smart contract interactions, and predicting network congestion. The results showcase the accuracy and efficiency of our approach, emphasizing its practical applicability in real-world scenarios. Security and Robustness: To ensure the reliability of our methodology, we conduct thorough security and robustness evaluations. Our approach demonstrates high resilience against adversarial attacks and perturbations, reaffirming its suitability for security-critical blockchain applications. Conclusion: By integrating graph-based data representation, GCNs, probabilistic sampling, and distributed computing, we achieve network scalability without compromising analytical precision. This approach addresses the pressing challenges posed by the expanding Ethereum network, opening new avenues for research and enabling real-time insights into decentralized ecosystems. Our work contributes to the development of scalable blockchain analytics, laying the foundation for sustainable growth and advancement in the domain of blockchain research and application.

Keywords: Ethereum, scalable network, GCN, probabilistic sampling, distributed computing

Procedia PDF Downloads 42

Authors: Jingling Li, Yi Zhang, Wei Liang, Tao Cui, Jun Li

Abstract:

Compared with terrestrial network, the traffic of spatial information network has both self-similarity and short correlation characteristics. By studying its traffic prediction method, the resource utilization of spatial information network can be improved, and the method can provide an important basis for traffic planning of a spatial information network. In this paper, considering the accuracy and complexity of the algorithm, the spatial information network traffic is decomposed into approximate component with long correlation and detail component with short correlation, and a time series hybrid prediction model based on wavelet decomposition is proposed to predict the spatial network traffic. Firstly, the original traffic data are decomposed to approximate components and detail components by using wavelet decomposition algorithm. According to the autocorrelation and partial correlation smearing and truncation characteristics of each component, the corresponding model (AR/MA/ARMA) of each detail component can be directly established, while the type of approximate component modeling can be established by ARIMA model after smoothing. Finally, the prediction results of the multiple models are fitted to obtain the prediction results of the original data. The method not only considers the self-similarity of a spatial information network, but also takes into account the short correlation caused by network burst information, which is verified by using the measured data of a certain back bone network released by the MAWI working group in 2018. Compared with the typical time series model, the predicted data of hybrid model is closer to the real traffic data and has a smaller relative root means square error, which is more suitable for a spatial information network.

Keywords: spatial information network, traffic prediction, wavelet decomposition, time series model

Procedia PDF Downloads 115