Search results for: finite element model/COMSOL multiphysics
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 19192

Search results for: finite element model/COMSOL multiphysics

12022 EQMamba - Method Suggestion for Earthquake Detection and Phase Picking

Authors: Noga Bregman

Abstract:

Accurate and efficient earthquake detection and phase picking are crucial for seismic hazard assessment and emergency response. This study introduces EQMamba, a deep-learning method that combines the strengths of the Earthquake Transformer and the Mamba model for simultaneous earthquake detection and phase picking. EQMamba leverages the computational efficiency of Mamba layers to process longer seismic sequences while maintaining a manageable model size. The proposed architecture integrates convolutional neural networks (CNNs), bidirectional long short-term memory (BiLSTM) networks, and Mamba blocks. The model employs an encoder composed of convolutional layers and max pooling operations, followed by residual CNN blocks for feature extraction. Mamba blocks are applied to the outputs of BiLSTM blocks, efficiently capturing long-range dependencies in seismic data. Separate decoders are used for earthquake detection, P-wave picking, and S-wave picking. We trained and evaluated EQMamba using a subset of the STEAD dataset, a comprehensive collection of labeled seismic waveforms. The model was trained using a weighted combination of binary cross-entropy loss functions for each task, with the Adam optimizer and a scheduled learning rate. Data augmentation techniques were employed to enhance the model's robustness. Performance comparisons were conducted between EQMamba and the EQTransformer over 20 epochs on this modest-sized STEAD subset. Results demonstrate that EQMamba achieves superior performance, with higher F1 scores and faster convergence compared to EQTransformer. EQMamba reached F1 scores of 0.8 by epoch 5 and maintained higher scores throughout training. The model also exhibited more stable validation performance, indicating good generalization capabilities. While both models showed lower accuracy in phase-picking tasks compared to detection, EQMamba's overall performance suggests significant potential for improving seismic data analysis. The rapid convergence and superior F1 scores of EQMamba, even on a modest-sized dataset, indicate promising scalability for larger datasets. This study contributes to the field of earthquake engineering by presenting a computationally efficient and accurate method for simultaneous earthquake detection and phase picking. Future work will focus on incorporating Mamba layers into the P and S pickers and further optimizing the architecture for seismic data specifics. The EQMamba method holds the potential for enhancing real-time earthquake monitoring systems and improving our understanding of seismic events.

Keywords: earthquake, detection, phase picking, s waves, p waves, transformer, deep learning, seismic waves

Procedia PDF Downloads 58
12021 Heuristic Algorithms for Time Based Weapon-Target Assignment Problem

Authors: Hyun Seop Uhm, Yong Ho Choi, Ji Eun Kim, Young Hoon Lee

Abstract:

Weapon-target assignment (WTA) is a problem that assigns available launchers to appropriate targets in order to defend assets. Various algorithms for WTA have been developed over past years for both in the static and dynamic environment (denoted by SWTA and DWTA respectively). Due to the problem requirement to be solved in a relevant computational time, WTA has suffered from the solution efficiency. As a result, SWTA and DWTA problems have been solved in the limited situation of the battlefield. In this paper, the general situation under continuous time is considered by Time based Weapon Target Assignment (TWTA) problem. TWTA are studied using the mixed integer programming model, and three heuristic algorithms; decomposed opt-opt, decomposed opt-greedy, and greedy algorithms are suggested. Although the TWTA optimization model works inefficiently when it is characterized by a large size, the decomposed opt-opt algorithm based on the linearization and decomposition method extracted efficient solutions in a reasonable computation time. Because the computation time of the scheduling part is too long to solve by the optimization model, several algorithms based on greedy is proposed. The models show lower performance value than that of the decomposed opt-opt algorithm, but very short time is needed to compute. Hence, this paper proposes an improved method by applying decomposition to TWTA, and more practical and effectual methods can be developed for using TWTA on the battlefield.

Keywords: air and missile defense, weapon target assignment, mixed integer programming, piecewise linearization, decomposition algorithm, military operations research

Procedia PDF Downloads 337
12020 The Influence of Gossip on the Absorption Probabilities in Moran Process

Authors: Jurica Hižak

Abstract:

Getting to know the agents, i.e., identifying the free riders in a population, can be considered one of the main challenges in establishing cooperation. An ordinary memory-one agent such as Tit-for-tat may learn “who is who” in the population through direct interactions. Past experiences serve them as a landmark to know with whom to cooperate and against whom to retaliate in the next encounter. However, this kind of learning is risky and expensive. A cheaper and less painful way to detect free riders may be achieved by gossiping. For this reason, as part of this research, a special type of Tit-for-tat agent was designed – a “Gossip-Tit-for-tat” agent that can share data with other agents of its kind. The performances of both strategies, ordinary Tit-for-tat and Gossip-Tit-for-tat, against Always-defect have been compared in the finite-game framework of the Iterated Prisoner’s Dilemma via the Moran process. Agents were able to move in a random-walk fashion, and they were programmed to play Prisoner’s Dilemma each time they met. Moreover, at each step, one randomly selected individual was eliminated, and one individual was reproduced in accordance with the Moran process of selection. In this way, the size of the population always remained the same. Agents were selected for reproduction via the roulette wheel rule, i.e., proportionally to the relative fitness of the strategy. The absorption probability was calculated after the population had been absorbed completely by cooperators, which means that all the states have been occupied and all of the transition probabilities have been determined. It was shown that gossip increases absorption probabilities and therefore enhances the evolution of cooperation in the population.

Keywords: cooperation, gossip, indirect reciprocity, Moran process, prisoner’s dilemma, tit-for-tat

Procedia PDF Downloads 99
12019 Experimental Determination of Aluminum 7075-T6 Parameters Using Stabilized Cycle Tests to Predict Thermal Ratcheting

Authors: Armin Rahmatfam, Mohammad Zehsaz, Farid Vakili Tahami, Nasser Ghassembaglou

Abstract:

In this paper the thermal ratcheting, kinematic hardening parameters C, γ, isotropic hardening parameters and also k, b, Q combined isotropic/kinematic hardening parameters have been obtained experimentally from the monotonic, strain controlled cyclic tests at room and elevated temperatures of 20°C, 100°C, and 400°C. These parameters are used in nonlinear combined isotropic/kinematic hardening model to predict better description of the loading and reloading cycles in the cyclic indentation as well as thermal ratcheting. For this purpose, three groups of specimens made of Aluminum 7075-T6 have been investigated. After each test and using stable hysteretic cycles, material parameters have been obtained for using in combined nonlinear isotropic/kinematic hardening models. Also the methodology of obtaining the correct kinematic/isotropic hardening parameters is presented.

Keywords: combined hardening model, kinematic hardening, isotropic hardening, cyclic tests

Procedia PDF Downloads 481
12018 Nonparametric Estimation of Risk-Neutral Densities via Empirical Esscher Transform

Authors: Manoel Pereira, Alvaro Veiga, Camila Epprecht, Renato Costa

Abstract:

This paper introduces an empirical version of the Esscher transform for risk-neutral option pricing. Traditional parametric methods require the formulation of an explicit risk-neutral model and are operational only for a few probability distributions for the returns of the underlying. In our proposal, we make only mild assumptions on the pricing kernel and there is no need for the formulation of the risk-neutral model for the returns. First, we simulate sample paths for the returns under the physical distribution. Then, based on the empirical Esscher transform, the sample is reweighted, giving rise to a risk-neutralized sample from which derivative prices can be obtained by a weighted sum of the options pay-offs in each path. We compare our proposal with some traditional parametric pricing methods in four experiments with artificial and real data.

Keywords: esscher transform, generalized autoregressive Conditional Heteroscedastic (GARCH), nonparametric option pricing

Procedia PDF Downloads 490
12017 Stock Prediction and Portfolio Optimization Thesis

Authors: Deniz Peksen

Abstract:

This thesis aims to predict trend movement of closing price of stock and to maximize portfolio by utilizing the predictions. In this context, the study aims to define a stock portfolio strategy from models created by using Logistic Regression, Gradient Boosting and Random Forest. Recently, predicting the trend of stock price has gained a significance role in making buy and sell decisions and generating returns with investment strategies formed by machine learning basis decisions. There are plenty of studies in the literature on the prediction of stock prices in capital markets using machine learning methods but most of them focus on closing prices instead of the direction of price trend. Our study differs from literature in terms of target definition. Ours is a classification problem which is focusing on the market trend in next 20 trading days. To predict trend direction, fourteen years of data were used for training. Following three years were used for validation. Finally, last three years were used for testing. Training data are between 2002-06-18 and 2016-12-30 Validation data are between 2017-01-02 and 2019-12-31 Testing data are between 2020-01-02 and 2022-03-17 We determine Hold Stock Portfolio, Best Stock Portfolio and USD-TRY Exchange rate as benchmarks which we should outperform. We compared our machine learning basis portfolio return on test data with return of Hold Stock Portfolio, Best Stock Portfolio and USD-TRY Exchange rate. We assessed our model performance with the help of roc-auc score and lift charts. We use logistic regression, Gradient Boosting and Random Forest with grid search approach to fine-tune hyper-parameters. As a result of the empirical study, the existence of uptrend and downtrend of five stocks could not be predicted by the models. When we use these predictions to define buy and sell decisions in order to generate model-based-portfolio, model-based-portfolio fails in test dataset. It was found that Model-based buy and sell decisions generated a stock portfolio strategy whose returns can not outperform non-model portfolio strategies on test dataset. We found that any effort for predicting the trend which is formulated on stock price is a challenge. We found same results as Random Walk Theory claims which says that stock price or price changes are unpredictable. Our model iterations failed on test dataset. Although, we built up several good models on validation dataset, we failed on test dataset. We implemented Random Forest, Gradient Boosting and Logistic Regression. We discovered that complex models did not provide advantage or additional performance while comparing them with Logistic Regression. More complexity did not lead us to reach better performance. Using a complex model is not an answer to figure out the stock-related prediction problem. Our approach was to predict the trend instead of the price. This approach converted our problem into classification. However, this label approach does not lead us to solve the stock prediction problem and deny or refute the accuracy of the Random Walk Theory for the stock price.

Keywords: stock prediction, portfolio optimization, data science, machine learning

Procedia PDF Downloads 81
12016 Towards an African Model: A Survey of Social Enterprises in South Africa

Authors: Kerryn Krige, Kerrin Myers

Abstract:

Social entrepreneurship offers the opportunity to simultaneously address both social and economic inequality in South Africa. Its appeal across racial groups, its attractiveness to young people, its applicability in rural and peri-urban markets, and its acceleration in middle income, large-business economies suits the South African context. However, the potential to deliver much-needed developmental benefits has not been realised because the social entrepreneurship debate lacks evidence as to who social entrepreneurs are, their goals and operations and the socio-economic results they achieve. As a result, policy development has been stunted, and legislative barriers and red tape remain. Social entrepreneurs are isolated from the mainstream economy, and struggle to access funding because of limitations in legislative and organisational structures. The objective of the study is to strengthen the ecosystem for social entrepreneurship in South Africa by producing robust, policy-rich information from and about social enterprises currently in operation across the country. The study employs a quantitative survey methodology, using online and telephonic data collection methods. A purposive sample of 1000 social enterprises was included in the first large-scale study of social entrepreneurship in South Africa. The results offer deep insight into the characteristics of social enterprises; the activities they undertake and the markets they serve; their modes of operation and funding sources as well as key challenges and support systems. The results contribute towards developing a model of social enterprise in the African context.

Keywords: social enterprise, key characteristics, challenges and enablers, towards an African model

Procedia PDF Downloads 310
12015 Speech Detection Model Based on Deep Neural Networks Classifier for Speech Emotions Recognition

Authors: Aisultan Shoiynbek, Darkhan Kuanyshbay, Paulo Menezes, Akbayan Bekarystankyzy, Assylbek Mukhametzhanov, Temirlan Shoiynbek

Abstract:

Speech emotion recognition (SER) has received increasing research interest in recent years. It is a common practice to utilize emotional speech collected under controlled conditions recorded by actors imitating and artificially producing emotions in front of a microphone. There are four issues related to that approach: emotions are not natural, meaning that machines are learning to recognize fake emotions; emotions are very limited in quantity and poor in variety of speaking; there is some language dependency in SER; consequently, each time researchers want to start work with SER, they need to find a good emotional database in their language. This paper proposes an approach to create an automatic tool for speech emotion extraction based on facial emotion recognition and describes the sequence of actions involved in the proposed approach. One of the first objectives in the sequence of actions is the speech detection issue. The paper provides a detailed description of the speech detection model based on a fully connected deep neural network for Kazakh and Russian. Despite the high results in speech detection for Kazakh and Russian, the described process is suitable for any language. To investigate the working capacity of the developed model, an analysis of speech detection and extraction from real tasks has been performed.

Keywords: deep neural networks, speech detection, speech emotion recognition, Mel-frequency cepstrum coefficients, collecting speech emotion corpus, collecting speech emotion dataset, Kazakh speech dataset

Procedia PDF Downloads 28
12014 Trigonelline: A Promising Compound for The Treatment of Alzheimer's Disease

Authors: Mai M. Farid, Ximeng Yang, Tomoharu Kuboyama, Chihiro Tohda

Abstract:

Trigonelline is a major alkaloid component derived from Trigonella foenum-graecum L. (fenugreek) and has been reported before as a potential neuroprotective agent, especially in Alzheimer’s disease (AD). However, the previous data were unclear and used model mice were not well established. In the present study, the effect of trigonelline on memory function was investigated in Alzheimer’s disease transgenic model mouse, 5XFAD which overexpresses the mutated APP and PS1 genes. Oral administration of trigonelline for 14 days significantly enhanced object recognition and object location memories. Plasma and cerebral cortex were isolated at 30 min, 1h, 3h, and 6 h after oral administration of trigonelline. LC-MS/MS analysis indicated that trigonelline was detected in both plasma and cortex from 30 min after, suggesting good penetration of trigonelline into the brain. In addition, trigonelline significantly ameliorated axonal and dendrite atrophy in Amyloid β-treated cortical neurons. These results suggest that trigonelline could be a promising therapeutic candidate for AD.

Keywords: alzheimer’s disease, cortical neurons, LC-MS/MS analysis, trigonelline

Procedia PDF Downloads 149
12013 Knowledge Transfer and the Translation of Technical Texts

Authors: Ahmed Alaoui

Abstract:

This paper contributes to the ongoing debate as to the relevance of translation studies to professional practitioners. It exposes the various misconceptions permeating the links between theory and practice in the translation landscape in the Arab World. It is a thesis of this paper that specialization in translation should be redefined; taking account of the fact, that specialized knowledge alone is neither crucial nor sufficient in technical translation. It should be tested against the readability of the translated text, the appropriateness of its style and the usability of its content by end-users to carry out their intended tasks. The paper also proposes a preliminary model to establish a working link between theory and practice from the perspective of professional trainers and practitioners, calling for the latter to participate in the production of knowledge in a systematic fashion. While this proposal is driven by a rather intuitive conviction, a research line is needed to specify the methodological moves to establish the mediation strategies that would relate the components in the model of knowledge transfer proposed in this paper.

Keywords: knowledge transfer, misconceptions, specialized texts, translation theory, translation practice

Procedia PDF Downloads 396
12012 Dynamic of Nonlinear Duopoly Game with Heterogeneous Players

Authors: Jixiang Zhang, Yanhua Wang

Abstract:

A dynamic of Bertrand duopoly game is analyzed, where players use different production methods and choose their prices with bounded rationality. The equilibriums of the corresponding discrete dynamical systems are investigated. The stability conditions of Nash equilibrium under a local adjustment process are studied. The stability conditions of Nash equilibrium under a local adjustment process are studied. The stability of Nash equilibrium, as some parameters of the model are varied, gives rise to complex dynamics such as cycles of higher order and chaos. On this basis, we discover that an increase of adjustment speed of bounded rational player can make Bertrand market sink into the chaotic state. Finally, the complex dynamics, bifurcations and chaos are displayed by numerical simulation.

Keywords: Bertrand duopoly model, discrete dynamical system, heterogeneous expectations, nash equilibrium

Procedia PDF Downloads 417
12011 Predicting National Football League (NFL) Match with Score-Based System

Authors: Marcho Setiawan Handok, Samuel S. Lemma, Abdoulaye Fofana, Naseef Mansoor

Abstract:

This paper is proposing a method to predict the outcome of the National Football League match with data from 2019 to 2022 and compare it with other popular models. The model uses open-source statistical data of each team, such as passing yards, rushing yards, fumbles lost, and scoring. Each statistical data has offensive and defensive. For instance, a data set of anticipated values for a specific matchup is created by comparing the offensive passing yards obtained by one team to the defensive passing yards given by the opposition. We evaluated the model’s performance by contrasting its result with those of established prediction algorithms. This research is using a neural network to predict the score of a National Football League match and then predict the winner of the game.

Keywords: game prediction, NFL, football, artificial neural network

Procedia PDF Downloads 85
12010 Digital Economy as an Alternative for Post-Pandemic Recovery in Latin America: A Literature Review

Authors: Armijos-Orellana Ana, González-Calle María, Maldonado-Matute Juan, Guerrero-Maxi Pedro

Abstract:

Nowadays, the digital economy represents a fundamental element to guarantee economic and social development, whose importance increased significantly with the arrival of the COVID-19 pandemic. However, despite the benefits it offers, it can also be detrimental to those developing countries characterized by a wide digital divide. It is for this reason that the objective of this research was to identify and describe the main characteristics, benefits, and obstacles of the digital economy for Latin American countries. Through a bibliographic review, using the analytical-synthetic method in the period 1995-2021, it was determined that the digital economy could give way to structural changes, reduce inequality, and promote processes of social inclusion, as well as promote the construction and participatory development of organizational structures and institutional capacities in Latin American countries. However, the results showed that the digital economy is still incipient in the region and at least three factors are needed to establish it: joint work between academia, the business sector and the State, greater emphasis on learning and application of digital transformation and the creation of policies that encourage the creation of digital organizations.

Keywords: developing countries, digital divide, digital economy, digital literacy, digital transformation

Procedia PDF Downloads 142
12009 CyberSecurity Malaysia: Towards Becoming a National Certification Body for Information Security Management Systems Internal Auditors

Authors: M. S. Razana, Z. W. Shafiuddin

Abstract:

Internal auditing is one of the most important activities for organizations that implement information security management systems (ISMS). The purpose of internal audits is to ensure the ISMS implementation is in accordance to the ISO/IEC 27001 standard and the organization’s own requirements for its ISMS. Competent internal auditors are the main element that contributes to the effectiveness of internal auditing activities. To realize this need, CyberSecurity Malaysia is now in the process of becoming a certification body that certifies ISMS internal auditors. The certification scheme will assess the competence of internal auditors in generic knowledge and skills in management systems, and also in ISMS-specific knowledge and skills. The certification assessment is based on the ISO/IEC 19011 Guidelines for auditing management systems, ISO/IEC 27007 Guidelines for information security management systems auditing and ISO/IEC 27001 Information security management systems requirements. The certification scheme complies with the ISO/IEC 17024 General requirements for bodies operating certification systems of persons. Candidates who pass the exam will be certified as an ISMS Internal Auditor, whose competency will be evaluated every three years.

Keywords: ISMS internal audit, ISMS internal auditor, ISO/IEC 17024, competence, certification

Procedia PDF Downloads 235
12008 The Data Quality Model for the IoT based Real-time Water Quality Monitoring Sensors

Authors: Rabbia Idrees, Ananda Maiti, Saurabh Garg, Muhammad Bilal Amin

Abstract:

IoT devices are the basic building blocks of IoT network that generate enormous volume of real-time and high-speed data to help organizations and companies to take intelligent decisions. To integrate this enormous data from multisource and transfer it to the appropriate client is the fundamental of IoT development. The handling of this huge quantity of devices along with the huge volume of data is very challenging. The IoT devices are battery-powered and resource-constrained and to provide energy efficient communication, these IoT devices go sleep or online/wakeup periodically and a-periodically depending on the traffic loads to reduce energy consumption. Sometime these devices get disconnected due to device battery depletion. If the node is not available in the network, then the IoT network provides incomplete, missing, and inaccurate data. Moreover, many IoT applications, like vehicle tracking and patient tracking require the IoT devices to be mobile. Due to this mobility, If the distance of the device from the sink node become greater than required, the connection is lost. Due to this disconnection other devices join the network for replacing the broken-down and left devices. This make IoT devices dynamic in nature which brings uncertainty and unreliability in the IoT network and hence produce bad quality of data. Due to this dynamic nature of IoT devices we do not know the actual reason of abnormal data. If data are of poor-quality decisions are likely to be unsound. It is highly important to process data and estimate data quality before bringing it to use in IoT applications. In the past many researchers tried to estimate data quality and provided several Machine Learning (ML), stochastic and statistical methods to perform analysis on stored data in the data processing layer, without focusing the challenges and issues arises from the dynamic nature of IoT devices and how it is impacting data quality. A comprehensive review on determining the impact of dynamic nature of IoT devices on data quality is done in this research and presented a data quality model that can deal with this challenge and produce good quality of data. This research presents the data quality model for the sensors monitoring water quality. DBSCAN clustering and weather sensors are used in this research to make data quality model for the sensors monitoring water quality. An extensive study has been done in this research on finding the relationship between the data of weather sensors and sensors monitoring water quality of the lakes and beaches. The detailed theoretical analysis has been presented in this research mentioning correlation between independent data streams of the two sets of sensors. With the help of the analysis and DBSCAN, a data quality model is prepared. This model encompasses five dimensions of data quality: outliers’ detection and removal, completeness, patterns of missing values and checks the accuracy of the data with the help of cluster’s position. At the end, the statistical analysis has been done on the clusters formed as the result of DBSCAN, and consistency is evaluated through Coefficient of Variation (CoV).

Keywords: clustering, data quality, DBSCAN, and Internet of things (IoT)

Procedia PDF Downloads 142
12007 A Model Suggestion on Competitiveness and Sustainability of SMEs in Developing Countries

Authors: Ahmet Diken, Tahsin Karabulut

Abstract:

The factor which developing countries are in need is capital. Such countries make an effort to increase their income in order to meet their expenses for employment, infrastructure, superstructure investments, education, health and defense. The sole income of the countries is taxes collected from businesses. The businesses should drive profit and return in order to be able to toll. In a world where competition exists, different strategies may be followed by business in developing countries and they must specify their target markets. İn order to minimize cost and maximize profit, SMEs have to concentrate on target markets and select cost oriented strategy. In this study, a theoretical model is suggested that SME firms have to act as cluster between each other, and also must be optimal provider for large scale firms. SMEs’ policy must be supported by public. This relationship can benefit large scale firms to have brand over the world, and this organization increases value added for developing countries.

Keywords: competitiveness, countries, SMEs developing, sustainability

Procedia PDF Downloads 316
12006 The Analysis of Swales Model (Cars Model) in the UMT Final Year Engineering Students

Authors: Kais Amir Kadhim

Abstract:

Context: The study focuses on the rhetorical structure of chapters in engineering final year projects, specifically the Introduction chapter, written by UMT (University of Marine Technology) engineering students. Existing research has explored the use of genre-based approaches to analyze the writing of final year projects in various disciplines. Research Aim: The aim of this study is to investigate the rhetorical structure of Introduction chapters in engineering final year projects by UMT students. The study aims to identify the frequency of communicative moves and their constituent steps within the Introduction chapters, as well as understand how students justify their research projects. Methodology: The research design will utilize a mixed method approach, combining both quantitative and qualitative methods. Forty Introduction chapters from two different fields in UMT engineering undergraduate programs will be selected for analysis. Findings: The study intends to identify the types of moves present in the Introduction chapters of engineering final year projects by UMT students. Additionally, it aims to determine if these moves and steps are obligatory, conventional, or optional. Theoretical Importance: The study draws upon Bunton's modified CARS (Creating a Research Space) model, which is a conceptual framework used for analyzing the introduction sections of theses. By applying this model, the study contributes to the understanding of the rhetorical structure of Introduction chapters in engineering final year projects. Data Collection: The study will collect data from forty Introduction chapters of engineering final year projects written by UMT engineering students. These chapters will be selected from two different fields within UMT's engineering undergraduate programs. Analysis Procedures: The analysis will involve identifying and categorizing the communicative moves and their constituent steps within the Introduction chapters. The study will utilize both quantitative and qualitative analysis methods to examine the frequency and nature of these moves. Question Addressed: The study aims to address the question of how UMT engineering students structure and justify their research projects within the Introduction chapters of their final year projects. Conclusion: The study aims to contribute to the knowledge of rhetorical structure in engineering final year projects by investigating the Introduction chapters written by UMT engineering students. By using a mixed method research design and applying the modified CARS model, the study intends to identify the types of moves and steps employed by students and explore their justifications for their research projects. The findings have the potential to enhance the understanding of effective academic writing in engineering disciplines.

Keywords: cohesive markers, learning, meaning, students

Procedia PDF Downloads 76
12005 Deep Neural Network Approach for Navigation of Autonomous Vehicles

Authors: Mayank Raj, V. G. Narendra

Abstract:

Ever since the DARPA challenge on autonomous vehicles in 2005, there has been a lot of buzz about ‘Autonomous Vehicles’ amongst the major tech giants such as Google, Uber, and Tesla. Numerous approaches have been adopted to solve this problem, which can have a long-lasting impact on mankind. In this paper, we have used Deep Learning techniques and TensorFlow framework with the goal of building a neural network model to predict (speed, acceleration, steering angle, and brake) features needed for navigation of autonomous vehicles. The Deep Neural Network has been trained on images and sensor data obtained from the comma.ai dataset. A heatmap was used to check for correlation among the features, and finally, four important features were selected. This was a multivariate regression problem. The final model had five convolutional layers, followed by five dense layers. Finally, the calculated values were tested against the labeled data, where the mean squared error was used as a performance metric.

Keywords: autonomous vehicles, deep learning, computer vision, artificial intelligence

Procedia PDF Downloads 160
12004 Statistical Comparison of Ensemble Based Storm Surge Forecasting Models

Authors: Amin Salighehdar, Ziwen Ye, Mingzhe Liu, Ionut Florescu, Alan F. Blumberg

Abstract:

Storm surge is an abnormal water level caused by a storm. Accurate prediction of a storm surge is a challenging problem. Researchers developed various ensemble modeling techniques to combine several individual forecasts to produce an overall presumably better forecast. There exist some simple ensemble modeling techniques in literature. For instance, Model Output Statistics (MOS), and running mean-bias removal are widely used techniques in storm surge prediction domain. However, these methods have some drawbacks. For instance, MOS is based on multiple linear regression and it needs a long period of training data. To overcome the shortcomings of these simple methods, researchers propose some advanced methods. For instance, ENSURF (Ensemble SURge Forecast) is a multi-model application for sea level forecast. This application creates a better forecast of sea level using a combination of several instances of the Bayesian Model Averaging (BMA). An ensemble dressing method is based on identifying best member forecast and using it for prediction. Our contribution in this paper can be summarized as follows. First, we investigate whether the ensemble models perform better than any single forecast. Therefore, we need to identify the single best forecast. We present a methodology based on a simple Bayesian selection method to select the best single forecast. Second, we present several new and simple ways to construct ensemble models. We use correlation and standard deviation as weights in combining different forecast models. Third, we use these ensembles and compare with several existing models in literature to forecast storm surge level. We then investigate whether developing a complex ensemble model is indeed needed. To achieve this goal, we use a simple average (one of the simplest and widely used ensemble model) as benchmark. Predicting the peak level of Surge during a storm as well as the precise time at which this peak level takes place is crucial, thus we develop a statistical platform to compare the performance of various ensemble methods. This statistical analysis is based on root mean square error of the ensemble forecast during the testing period and on the magnitude and timing of the forecasted peak surge compared to the actual time and peak. In this work, we analyze four hurricanes: hurricanes Irene and Lee in 2011, hurricane Sandy in 2012, and hurricane Joaquin in 2015. Since hurricane Irene developed at the end of August 2011 and hurricane Lee started just after Irene at the beginning of September 2011, in this study we consider them as a single contiguous hurricane event. The data set used for this study is generated by the New York Harbor Observing and Prediction System (NYHOPS). We find that even the simplest possible way of creating an ensemble produces results superior to any single forecast. We also show that the ensemble models we propose generally have better performance compared to the simple average ensemble technique.

Keywords: Bayesian learning, ensemble model, statistical analysis, storm surge prediction

Procedia PDF Downloads 309
12003 Automated Weight Painting: Using Deep Neural Networks to Adjust 3D Mesh Skeletal Weights

Authors: John Gibbs, Benjamin Flanders, Dylan Pozorski, Weixuan Liu

Abstract:

Weight Painting–adjusting the influence a skeletal joint has on a given vertex in a character mesh–is an arduous and time con- suming part of the 3D animation pipeline. This process generally requires a trained technical animator and many hours of work to complete. Our skiNNer plug-in, which works within Autodesk’s Maya 3D animation software, uses Machine Learning and data pro- cessing techniques to create a deep neural network model that can accomplish the weight painting task in seconds rather than hours for bipedal quasi-humanoid character meshes. In order to create a properly trained network, a number of challenges were overcome, including curating an appropriately large data library, managing an arbitrary 3D mesh size, handling arbitrary skeletal architectures, accounting for extreme numeric values (most data points are near 0 or 1 for weight maps), and constructing an appropriate neural network model that can properly capture the high frequency alter- ation between high weight values (near 1.0) and low weight values (near 0.0). The arrived at neural network model is a cross between a traditional CNN, deep residual network, and fully dense network. The resultant network captures the unusually hard-edged features of a weight map matrix, and produces excellent results on many bipedal models.

Keywords: 3d animation, animation, character, rigging, skinning, weight painting, machine learning, artificial intelligence, neural network, deep neural network

Procedia PDF Downloads 274
12002 Two-Channels Thermal Energy Storage Tank: Experiments and Short-Cut Modelling

Authors: M. Capocelli, A. Caputo, M. De Falco, D. Mazzei, V. Piemonte

Abstract:

This paper presents the experimental results and the related modeling of a thermal energy storage (TES) facility, ideated and realized by ENEA and realizing the thermocline with an innovative geometry. Firstly, the thermal energy exchange model of an equivalent shell & tube heat exchanger is described and tested to reproduce the performance of the spiral exchanger installed in the TES. Through the regression of the experimental data, a first-order thermocline model was also validated to provide an analytical function of the thermocline, useful for the performance evaluation and the comparison with other systems and implementation in simulations of integrated systems (e.g. power plants). The experimental data obtained from the plant start-up and the short-cut modeling of the system can be useful for the process analysis, for the scale-up of the thermal storage system and to investigate the feasibility of its implementation in actual case-studies.

Keywords: CSP plants, thermal energy storage, thermocline, mathematical modelling, experimental data

Procedia PDF Downloads 331
12001 Combining Multiscale Patterns of Weather and Sea States into a Machine Learning Classifier for Mid-Term Prediction of Extreme Rainfall in North-Western Mediterranean Sea

Authors: Pinel Sebastien, Bourrin François, De Madron Du Rieu Xavier, Ludwig Wolfgang, Arnau Pedro

Abstract:

Heavy precipitation constitutes a major meteorological threat in the western Mediterranean. Research has investigated the relationship between the states of the Mediterranean Sea and the atmosphere with the precipitation for short temporal windows. However, at a larger temporal scale, the precursor signals of heavy rainfall in the sea and atmosphere have drawn little attention. Moreover, despite ongoing improvements in numerical weather prediction, the medium-term forecasting of rainfall events remains a difficult task. Here, we aim to investigate the influence of early-spring environmental parameters on the following autumnal heavy precipitations. Hence, we develop a machine learning model to predict extreme autumnal rainfall with a 6-month lead time over the Spanish Catalan coastal area, based on i) the sea pattern (main current-LPC and Sea Surface Temperature-SST) at the mesoscale scale, ii) 4 European weather teleconnection patterns (NAO, WeMo, SCAND, MO) at synoptic scale, and iii) the hydrological regime of the main local river (Rhône River). The accuracy of the developed model classifier is evaluated via statistical analysis based on classification accuracy, logarithmic and confusion matrix by comparing with rainfall estimates from rain gauges and satellite observations (CHIRPS-2.0). Sensitivity tests are carried out by changing the model configuration, such as sea SST, sea LPC, river regime, and synoptic atmosphere configuration. The sensitivity analysis suggests a negligible influence from the hydrological regime, unlike SST, LPC, and specific teleconnection weather patterns. At last, this study illustrates how public datasets can be integrated into a machine learning model for heavy rainfall prediction and can interest local policies for management purposes.

Keywords: extreme hazards, sensitivity analysis, heavy rainfall, machine learning, sea-atmosphere modeling, precipitation forecasting

Procedia PDF Downloads 138
12000 Screening of Strategic Management Criterions in Hospitals Using Delphi-Fuzzy Method

Authors: Helia Moayedi, Mahdi Moaidi

Abstract:

Nowadays, the managing and planning of hospitals is facing many problems. Failure to recognize the main criteria for strategic management to ensure long-term hospital performance can lead to many health problems. To achieve this goal, a qualitative-quantitate method titled Delphi-Fuzzy has been applied. This strategy makes it possible for experts to screen among the most important criteria in strategic management. To conduct this operation, a statistical society consisting of 20 experts in Ahwaz hospitals has been questioned. The final model confirms the key criterions after three stages of Delphi. This model provides the possibility to focus on the basic criteria and can determine the organization’s main orientation.

Keywords: Delphi-fuzzy method, hospital management, long-term planning, qualitative-quantitate method, screening of strategic criteria, strategic planning

Procedia PDF Downloads 133
11999 Study Mercapto-Nanoscavenger as a Promising Analytical Tool

Authors: Mohammed M. Algaradah

Abstract:

A chelating mercapto- nanoscavenger has been developed exploiting the high surface area of monodisperse nano-sized mesoporous silica. The nanoscavenger acts as a solid phase trace metal extractant whilst suspended as a quasi-stable sol in aqueous samples. This mode of extraction requires no external agitation as the particles move naturally through the sample by Brownian motion, convection and slow sedimentation. Careful size selection enables the nanoscavenger to be easily recovered together with the extracted analyte by conventional filtration or centrifugation. The research describes the successful attachment of chelator mercapto to ca. 136 ± 15 nm high surface area (BET surface area = 1006 m2 g-1) mesoporous silica particles. The resulting material had a copper capacity of ca. 1.34 ± 0.10 mmol g-1 and was successfully applied to the collection of a trace element from water. Essentially complete recovery of Cu (II) has been achieved from freshwater samples giving typical preconcentration factors of 100 from 50 µg/l samples. Data obtained from a nanoscavenger-based extraction of copper from samples were not significantly different from those obtained by using a conventional colorimetric procedure employing complexation/solvent extraction.

Keywords: nano scavenger, mesoporous silica, trace metal, preconcentration

Procedia PDF Downloads 88
11998 The Next Frontier for Mobile Based Augmented Reality: An Evaluation of AR Uptake in India

Authors: K. Krishna Milan Rao, Nelvin Joseph, Praveen Dwarakanath

Abstract:

Augmented and Virtual Realties is quickly becoming a hotbed of activity with millions of dollars being spent on R & D and companies such as Google and Microsoft rushing to stake their claim. Augmented reality (AR) is however marching ahead due to the spread of the ideal AR device – the smartphone. Despite its potential, there remains a deep digital divide between the Developed and Developing Countries. The Technological Acceptance Model (TAM) and Hofstede cultural dimensions also predict the behaviour intention to uptake AR in India will be large. This paper takes a quantified approach by collecting 340 survey responses to AR scenarios and analyzing them through statistics. The Survey responses show that the Intention to Use, Perceived Usefulness and Perceived Enjoyment dimensions are high among the urban population in India. This along with the exponential smartphone indicates that India is on the cusp of a boom in the AR sector.

Keywords: mobile augmented reality, technology acceptance model, Hofstede, cultural dimensions, India

Procedia PDF Downloads 251
11997 Investigation on the Structure of Temperature-Responsive N-isopropylacrylamide Microgels Containing a New Hydrophobic Crosslinker

Authors: G. Roshan Deen, J. S. Pedersen

Abstract:

Temperature-responsive poly(N-isopropyl acrylamide) PNIPAM microgels crosslinked with a new hydrophobic chemical crosslinker was prepared by surfactant-mediated precipitation emulsion polymerization. The temperature-responsive property of the microgel and the influence of the crosslinker on the swelling behaviour was studied systematically by light scattering and small-angle X-ray scattering (SAXS). The radius of gyration (Rg) and the hydrodynamic radius (Rh) of the microgels decreased with increase in temperature due to the volume phase transition from a swollen to a collapsed state. The ratio of Rg/Rh below the transition temperature was lower than that of hard-spheres due to the lower crosslinking density of the microgels. The SAXS data was analysed by a model in which the microgels were modelled as core-shell particles with a graded interface. The model at intermediate temperatures included a central core and a more diffuse outer layer describing pending polymer chains with a low crosslinking density. In the fully swollen state, the microgels were modelled with a single component with a broad graded surface. In the collapsed state they were modelled as homogeneous and relatively compact particles. The polymer volume fraction inside the microgel was also derived based on the model and was found to increase with increase in temperature as a result of collapse of the microgel to compact particles. The polymer volume fraction in the core of the microgel in the collapsed state was about 60% which is higher than that of similar microgels crosslinked with hydrophilic and flexible cross-linkers.

Keywords: microgels, SAXS, hydrophobic crosslinker, light scattering

Procedia PDF Downloads 427
11996 Heuristic for Scheduling Correlated Parallel Machine to Minimize Maximum Lateness and Total Weighed Completion Time

Authors: Yang-Kuei Lin, Yun-Xi Zhang

Abstract:

This research focuses on the bicriteria correlated parallel machine scheduling problem. The two objective functions considered in this problem are to minimize maximum lateness and total weighted completion time. We first present a mixed integer programming (MIP) model that can find the entire efficient frontier for the studied problem. Next, we have proposed a bicriteria heuristic that can find non-dominated solutions for the studied problem. The performance of the proposed bicriteria heuristic is compared with the efficient frontier generated by solving the MIP model. Computational results indicate that the proposed bicriteria heuristic can solve the problem efficiently and find a set of diverse solutions that are uniformly distributed along the efficient frontier.

Keywords: bicriteria, correlated parallel machines, heuristic, scheduling

Procedia PDF Downloads 142
11995 Business Model Innovation and Firm Performance: Exploring Moderation Effects

Authors: Mohammad-Ali Latifi, Harry Bouwman

Abstract:

Changes in the business environment accelerated dramatically over the last decades as a result of changes in technology, regulation, market, and competitors’ behavior. Firms need to change the way they do business in order to survive or maintain their growth. Innovating business model (BM) can create competitive advantages and enhance firm performance. However, many companies fail to achieve expected outcomes in practice, mostly due to irreversible fundamental changes in key components of the company’s BM. This leads to more ambiguity, uncertainty, and risks associated with business performance. However, the relationship among BM Innovation, moderating factors, and the firm’s overall performance is by and large ignored in the current literature. In this study, we identified twenty moderating factors from our comprehensive literature review. We categorized these factors based on two criteria regarding the extent to which: the moderating factors can be controlled and managed by firms, and they are generic or specific changes to the firms. This leads to four moderation groups. The first group is BM implementation, which includes management support, employees’ commitment, employees’ skills, communication, detailed plan. The second group is called BM practices, which consists of BM tooling, BM experimentation, the scope of change, speed of change, degree of novelty. The third group is Firm characteristics, including firm size, age, and ownership. The last group is called Industry characteristics, which considers the industry sector, competitive intensity, industry life cycle, environmental dynamism, high-tech vs. low-tech industry. Through collecting data from 508 European small and medium-sized enterprises (SMEs) and using the structural equation modeling technique, the developed moderation model was examined. Results revealed that all factors highlighted through these four groups moderate the relation between BMI and firm performance significantly. Particularly, factors related to BM-Implementation and BM-Practices are more manageable and would potentially improve firm overall performance. We believe that this result is more important for researchers and practitioners since the possibility of working on factors in Firm characteristics and Industry characteristics groups are limited, and the firm can hardly control and manage them to improve the performance of BMI efforts.

Keywords: business model innovation, firm performance, implementation, moderation

Procedia PDF Downloads 123
11994 Synthesis and Thermoluminescence Study of Nanocrystalline Radiation Dosimeter CaSO₄:Ce/Sm/Dy

Authors: Anant Pandey, Kanika Sharma, Vibha Chopra, Shaila Bahl, Pratik Kumar, S. P. Lochab, Birendra Singh

Abstract:

This paper reports the thermoluminescence (TL) properties of nanocrystalline CaSO₄ activated by Ce, Sm, and Dy. TL properties are investigated by chiefly changing the dopant element and also by varying the concentration of the dopant elements (from 0.05 mol % to 0.5 mol %) so as to establish the optimized dopant concentration for each of the activators. The method of salt preparation used is the typical chemical co-precipitation method and the technique used for characterization of the prepared samples is the X-Ray Diffraction (XRD) technique. Further, the phosphors are irradiated with gamma radiation from Co-60 (1.25 MeV) source (dose range- 30 Gy to 500 Gy). The optimized concentration (vis-a-vis TL peak intensity) of activator for CaSO₄:Ce is found to be 0.2 mol %, for CaSO₄:Sm it is 0.1 mol % and for CaSO₄:Dy it is 0.2 mol %. Further, the primary study of the TL response curves for all the three phosphors confirms linearity in the studied dose range (i.e., 30 Gy to 500 Gy). Finally, CaSO₄:Dy was also studied for its energy dependence property which plays an important role in defining the utility of a phosphor for dosimetric applications. The range of doses used for the energy dependence study was from 30 Gy to 500 Gy from Cs-137 (0.662 MeV). The nano-phosphors showed potential to be used as radiation dosimeter in the studied range of gamma radiation and thus must be studied for a wider range of doses.

Keywords: gamma radiation, nanocrystalline, radiation dosimetry, thermoluminescence

Procedia PDF Downloads 176
11993 Guests’ Satisfaction and Intention to Revisit Smart Hotels: Qualitative Interviews Approach

Authors: Raymond Chi Fai Si Tou, Jacey Ja Young Choe, Amy Siu Ian So

Abstract:

Smart hotels can be defined as the hotel which has an intelligent system, through digitalization and networking which achieve hotel management and service information. In addition, smart hotels include high-end designs that integrate information and communication technology with hotel management fulfilling the guests’ needs and improving the quality, efficiency and satisfaction of hotel management. The purpose of this study is to identify appropriate factors that may influence guests’ satisfaction and intention to revisit Smart Hotels based on service quality measurement of lodging quality index and extended UTAUT theory. Unified Theory of Acceptance and Use of Technology (UTAUT) is adopted as a framework to explain technology acceptance and use. Since smart hotels are technology-based infrastructure hotels, UTATU theory could be as the theoretical background to examine the guests’ acceptance and use after staying in smart hotels. The UTAUT identifies four key drivers of the adoption of information systems: performance expectancy, effort expectancy, social influence, and facilitating conditions. The extended UTAUT modifies the definitions of the seven constructs for consideration; the four previously cited constructs of the UTAUT model together with three new additional constructs, which including hedonic motivation, price value and habit. Thus, the seven constructs from the extended UTAUT theory could be adopted to understand their intention to revisit smart hotels. The service quality model will also be adopted and integrated into the framework to understand the guests’ intention of smart hotels. There are rare studies to examine the service quality on guests’ satisfaction and intention to revisit in smart hotels. In this study, Lodging Quality Index (LQI) will be adopted to measure the service quality in smart hotels. Using integrated UTAUT theory and service quality model because technological applications and services require using more than one model to understand the complicated situation for customers’ acceptance of new technology. Moreover, an integrated model could provide more perspective insights to explain the relationships of the constructs that could not be obtained from only one model. For this research, ten in-depth interviews are planned to recruit this study. In order to confirm the applicability of the proposed framework and gain an overview of the guest experience of smart hotels from the hospitality industry, in-depth interviews with the hotel guests and industry practitioners will be accomplished. In terms of the theoretical contribution, it predicts that the integrated models from the UTAUT theory and the service quality will provide new insights to understand factors that influence the guests’ satisfaction and intention to revisit smart hotels. After this study identifies influential factors, smart hotel practitioners could understand which factors may significantly influence smart hotel guests’ satisfaction and intention to revisit. In addition, smart hotel practitioners could also provide outstanding guests experience by improving their service quality based on the identified dimensions from the service quality measurement. Thus, it will be beneficial to the sustainability of the smart hotels business.

Keywords: intention to revisit, guest satisfaction, qualitative interviews, smart hotels

Procedia PDF Downloads 209