Search results for: bit error probability

Authors: Qiang Wang, Hongyang Yu

Abstract:

Multi-human 3D pose estimation is a challenging task in computer vision, which aims to recover the 3D joint locations of multiple people from multi-view images. In contrast to traditional methods, which typically only use color (RGB) images as input, our approach utilizes both color and depth (D) information contained in RGB-D images. We also employ a transformer-based model as the backbone of our approach, which is able to capture long-range dependencies and has been shown to perform well on various sequence modeling tasks. Our method is trained and tested on the Carnegie Mellon University (CMU) Panoptic dataset, which contains a diverse set of indoor and outdoor scenes with multiple people in varying poses and clothing. We evaluate the performance of our model on the standard 3D pose estimation metrics of mean per-joint position error (MPJPE). Our results show that the transformer-based approach outperforms traditional methods and achieves competitive results on the CMU Panoptic dataset. We also perform an ablation study to understand the impact of different design choices on the overall performance of the model. In summary, our work demonstrates the effectiveness of using a transformer-based approach with RGB-D images for multi-human 3D pose estimation and has potential applications in real-world scenarios such as human-computer interaction, robotics, and augmented reality.

Keywords: multi-human 3D pose estimation, RGB-D images, transformer, 3D joint locations

Procedia PDF Downloads 61

Authors: D. Calogine, O. Chau, S. Dotti, O. Ramiarinjanahary, P. Rasoavonjy, F. Tovondahiniriko

Abstract:

Mafate is a natural circus in the north-western part of Reunion Island, without an electrical grid and road network. A micro-grid concept is being experimented in this area, composed of a photovoltaic production combined with electrochemical batteries, in order to meet the local population for self-consumption of electricity demands. This work develops a discrete model as well as a stochastic model in order to reach an optimal equilibrium between production and consumptions for a cluster of houses. The management of the energy power leads to a large linearized programming system, where the time interval of interest is 24 hours The experimental data are solar production, storage energy, and the parameters of the different electrical devices and batteries. The unknown variables to evaluate are the consumptions of the various electrical services, the energy drawn from and stored in the batteries, and the inhabitants’ planning wishes. The objective is to fit the solar production to the electrical consumption of the inhabitants, with an optimal use of the energies in the batteries by satisfying as widely as possible the users' planning requirements. In the discrete model, the different parameters and solutions of the linear programming system are deterministic scalars. Whereas in the stochastic approach, the data parameters and the linear programming solutions become random variables, then the distributions of which could be imposed or established by estimation from samples of real observations or from samples of optimal discrete equilibrium solutions.

Keywords: photovoltaic production, power consumption, battery storage resources, random variables, stochastic modeling, estimations of probability distributions, mixed integer linear programming, smart micro-grid, self-consumption of electricity.

Procedia PDF Downloads 91

Authors: Anas Rao, Hao Duan, Fanhua Ma

Abstract:

The dependency on the fossil fuels can be minimized by using the hydrogen enriched compressed natural gas (HCNG) in the transportation vehicles. However, the NOx emissions of HCNG engines are significantly higher, and this turned to be its major drawback. Therefore, the study of NOx emission of HCNG engines is a very important area of research. In this context, the experiments have been performed at the different hydrogen percentage, ignition timing, air-fuel ratio, manifold-absolute pressure, load and engine speed. Afterwards, the simulation has been accomplished by the quasi-dimensional combustion model of HCNG engine. In order to investigate the NOx emission, the NO mechanism has been coupled to the quasi-dimensional combustion model of HCNG engine. The three NOx mechanism: the thermal NOx, prompt NOx and N2O mechanism have been used to predict NOx emission. For the validation purpose, NO curve has been transformed into NO packets based on the temperature difference of 100 K for the lean-burn and 60 K for stoichiometric condition. While, the width of the packet has been taken as the ratio of crank duration of the packet to the total burnt duration. The combustion chamber of the engine has been divided into three zones, with the zone equal to the product of summation of NO packets and space. In order to check the accuracy of the model, the percentage error of NOx emission has been evaluated, and it lies in the range of ±6% and ±10% for the lean-burn and stoichiometric conditions respectively. Finally, the percentage contribution of each NO formation has been evaluated.

Keywords: quasi-dimensional combustion , thermal NO, prompt NO, NO packet

Procedia PDF Downloads 230

Authors: K. Swapna

Abstract:

A Series of Alkali-Earth Boro Tellurite (AEBT) glasses doped with different concentrations of Dy³⁺ ions have been prepared by using melt quenching technique and characterized through spectroscopic techniques such as optical absorption, excitation, emission and photoluminescence decay to understand their utility in optoelectronic devices such as lasers and white light emitting diodes (w-LEDs). Raman spectrum recorded for an undoped glass is used to measure the phonon energy of the host glass and various functional groups present in the host glass (AEBT). The intensities of the electronic transitions and the ligand environment around the Dy³⁺ ions were studied by applying Judd-Ofelt (J-O) theory to the recorded absorption spectra of the glasses. The evaluated J-O parameters are subsequently used to measure various radiative parameters such as transition probability (AR), radiative branching ratio (βR) and radiative lifetimes (τR) for the prominent fluorescent levels of Dy³⁺ ions in the as-prepared glasses. The luminescence spectra recorded at 387 nm excitation show three emission transitions (⁴F9/2→⁶H15/2 (blue), ⁴F9/2→⁶H13/2 (yellow) and ⁴F9/2 → ⁶H11/2 (red)) of which the yellow transition observed at 575 nm is found to be highly intense. The experimental branching ratio (βexp) and stimulated emission crosssection (σse) were measured from luminescence spectra. The experimental lifetimes (τexp) measured from the decay spectral profiles are combined with radiative lifetimes to measure quantum efficiencies of the as-prepared glasses. The yellow to blue intensity ratios and chromaticity color coordinates are found to vary with Dy³⁺ ion concentrations. The aforementioned results reveal that these glasses are aptly suitable for w-LEDs and laser devices.

Keywords: glasses, J-O parameters, photoluminescence, I-H model

Procedia PDF Downloads 132

Authors: Akalu Banbeta, Emmanuel Lesaffre, Reynaldo Martina, Joost Van Rosmalen

Abstract:

Including data from previous studies (historical data) in the analysis of the current study may reduce the sample size requirement and/or increase the power of analysis. The most common example is incorporating historical control data in the analysis of a current clinical trial. However, this only applies when the historical control dataare similar enough to the current control data. Recently, several Bayesian approaches for incorporating historical data have been proposed, such as the meta-analytic-predictive (MAP) prior and the modified power prior (MPP) both for single control as well as for multiple historical control arms. Here, we examine the performance of the MAP and the MPP approaches for the analysis of (over-dispersed) count data. To this end, we propose a computational method for the MPP approach for the Poisson and the negative binomial models. We conducted an extensive simulation study to assess the performance of Bayesian approaches. Additionally, we illustrate our approaches on an overactive bladder data set. For similar data across the control arms, the MPP approach outperformed the MAP approach with respect to thestatistical power. When the means across the control arms are different, the MPP yielded a slightly inflated type I error (TIE) rate, whereas the MAP did not. In contrast, when the dispersion parameters are different, the MAP gave an inflated TIE rate, whereas the MPP did not.We conclude that the MPP approach is more promising than the MAP approach for incorporating historical count data.

Keywords: count data, meta-analytic prior, negative binomial, poisson

Procedia PDF Downloads 99

Authors: Panupong Makvichian

Abstract:

Global Navigation Satellite System (GNSS) is nowadays a common technology that improves navigation functions in our life. Additionally, GNSS is also being employed on behalf of an accurate atmospheric sensor these times. Meteorology is a practical application of GNSS, which is unnoticeable in the background of people’s life. GNSS Precise Point Positioning (PPP) is a positioning method that requires data from a single dual-frequency receiver and precise information about satellite positions and satellite clocks. In addition, careful attention to mitigate various error sources is required. All the above data are combined in a sophisticated mathematical algorithm. At this point, the research is going to demonstrate how GNSS and PPP method is capable to provide high-precision estimates, such as 3D positions or Zenith tropospheric delays (ZTDs). ZTDs combined with pressure and temperature information allows us to estimate the water vapor in the atmosphere as precipitable water vapor (PWV). If the process is replicated for a network of GNSS sensors, we can create thematic maps that allow extract water content information in any location within the network area. All of the above are possible thanks to the advances in GNSS data processing. Therefore, we are able to use GNSS data for climatic trend analysis and acquisition of the further knowledge about the atmospheric water content.

Keywords: GNSS, precise point positioning, Zenith tropospheric delays, precipitable water vapor

Procedia PDF Downloads 178

Authors: Rinat Arbel-Goren, Joel Stavans

Abstract:

Cell-to-cell variations in transcript or protein abundance, called noise, may give rise to phenotypic variability between isogenic cells, enhancing the probability of survival under stress conditions. These variations may be introduced by post-transcriptional regulatory processes such as non-coding, small RNAs stoichiometric degradation of target transcripts in bacteria. We study the iron homeostasis network in Escherichia coli, in which the RyhB small RNA regulates the expression of various targets as a model system. Using fluorescence reporter genes to detect protein levels and single-molecule fluorescence in situ hybridization to monitor transcripts levels in individual cells, allows us to compare noise at both transcript and protein levels. The experimental results and computer simulations show that extrinsic noise buffers through a feed-forward loop configuration the increase in variability introduced at the transcript level by iron deprivation, illuminating the important role that extrinsic noise plays during stress. Surprisingly, extrinsic noise also decouples of fluctuations of two different targets, in spite of RyhB being a common upstream factor degrading both. Thus, phenotypic variability increases under stress conditions by the decoupling of target fluctuations in the same cell rather than by increasing the noise of each. We also present preliminary results on the adaptation of cells to prolonged iron deprivation in order to shed light on the evolutionary role of post-transcriptional downregulation by small RNAs.

Keywords: cell-to-cell variability, Escherichia coli, noise, single-molecule fluorescence in situ hybridization (smFISH), transcript

Procedia PDF Downloads 143

Authors: C. Kuoh, C. Lucas, T. Lopes, I. Mechie, J. Yoong, W. Yoong

Abstract:

For all surgeons, there is one preventable but still highly occurring complication – wrong site surgeries. They can have potentially catastrophic, irreversible, or even fatal consequences on patients. With the exponential development of microsurgery and the use of advanced technological tools, the consequences of operating on the wrong side, anatomical part, or even person is seen as the most visible and destructive of all surgical errors and perhaps the error that is dreaded by most clinicians as it threatens their licenses and arouses feelings of guilt. Despite the implementation of the WHO surgical safety checklist more than a decade ago, the incidence of wrong-site surgeries remains relatively high, leading to tremendous physical and psychological repercussions for the clinicians involved, as well as a financial burden for the healthcare institution. In this presentation, the authors explore various factors which can lead to wrong site surgery – a combination of environmental and human factors and evaluate their impact amongst patients, practitioners, their families, and the medical industry. Major contributing factors to these “never events” include deviations from checklists, excessive workload, and poor communication. Two real-life cases are discussed, and systems that can be implemented to prevent these errors are highlighted alongside lessons learnt from other industries. The authors suggest that reinforcing speaking-up, implementing medical professional trainings, and higher patient’s involvements can potentially improve safety in surgeries and electrosurgeries.

Keywords: wrong side surgery, never events, checklist, workload, communication

Procedia PDF Downloads 167

Authors: Jiao Sun, Li Pan, Shijun Liu

Abstract:

Recommender systems have been widely used in contemporary industry, and plenty of work has been done in this field to help users to identify items of interest. Collaborative Filtering (CF, for short) algorithm is an important technology in recommender systems. However, less work has been done in automobile recommendation system with the sharp increase of the amount of automobiles. What’s more, the computational speed is a major weakness for collaborative filtering technology. Therefore, using MapReduce framework to optimize the CF algorithm is a vital solution to this performance problem. In this paper, we present a recommendation of the users’ comment on industrial automobiles with various properties based on real world industrial datasets of user-automobile comment data collection, and provide recommendation for automobile providers and help them predict users’ comment on automobiles with new-coming property. Firstly, we solve the sparseness of matrix using previous construction of score matrix. Secondly, we solve the data normalization problem by removing dimensional effects from the raw data of automobiles, where different dimensions of automobile properties bring great error to the calculation of CF. Finally, we use the MapReduce framework to optimize the CF algorithm, and the computational speed has been improved times. UV decomposition used in this paper is an often used matrix factorization technology in CF algorithm, without calculating the interpolation weight of neighbors, which will be more convenient in industry.

Keywords: collaborative filtering, recommendation, data normalization, mapreduce

Procedia PDF Downloads 202

Authors: Hicham Gourgue, Ahmed Aharoune, Ahmed Ihlal

Abstract:

We present in this work our model of road traffic emissions (line sources) and dispersion of these emissions, named DISPOLSPEM (Dispersion of Poly Sources and Pollutants Emission Model). In its emission part, this model was designed to keep the consistent bottom-up and top-down approaches. It also allows to generate emission inventories from reduced input parameters being adapted to existing conditions in Morocco and in the other developing countries. While several simplifications are made, all the performance of the model results are kept. A further important advantage of the model is that it allows the uncertainty calculation and emission rate uncertainty according to each of the input parameters. In the dispersion part of the model, an improved line source model has been developed, implemented and tested against a reference solution. It provides improvement in accuracy over previous formulas of line source Gaussian plume model, without being too demanding in terms of computational resources. In the case study presented here, the biggest errors were associated with the ends of line source sections; these errors will be canceled by adjacent sections of line sources during the simulation of a road network. In cases where the wind is parallel to the source line, the use of the combination discretized source and analytical line source formulas minimizes remarkably the error. Because this combination is applied only for a small number of wind directions, it should not excessively increase the calculation time.

Keywords: air pollution, dispersion, emissions, line sources, road traffic, urban transport

Procedia PDF Downloads 423

Authors: Mahshid Asadollahi, Mohammad Akbari Asl

Abstract:

The restaurant industry is highly competitive, and restaurants are constantly looking for ways to attract new customers and retain their existing ones. Nonverbal communication is an important factor in creating a positive customer experience and can play a significant role in attracting customers to restaurants. Nonverbal communication can include body language, facial expressions, tone of voice, and physical proximity, among other things. The present study aimed to investigate the impact of nonverbal communication between restaurant employees and customers on attracting customers in food courts in Tehran. The research method was descriptive-correlational, and the statistical population of this study included all customers of food court restaurants in Tehran, which was about 30 restaurants. The research sample was selected through probability sampling, and 440 customers completed emotional response, customer satisfaction, and nonverbal communication questionnaires in person. The data obtained were analyzed using multiple regression analysis. The results showed that vocal language, employee proximity, physical appearance, and speech movements, as components of nonverbal communication of restaurant employees, had an impact on attracting customers. Additionally, positive and negative emotions of customers have a significant relationship with customer attraction in Food Court restaurants. The study shows that various nonverbal communication factors can play a significant role in attracting customers, and that positive and negative customer emotions can affect customer satisfaction. Therefore, restaurant owners and managers should pay attention to nonverbal communication and train their employees accordingly to create a positive and welcoming atmosphere for customers.

Keywords: verbal language, proximity of employees, physical appearance, speech gestures, nonverbal communication, customer emotions, customer attraction

Procedia PDF Downloads 74

Authors: Spandan Sahu, Amiya Kumar Pati, Kishanjit Kumar Khatua

Abstract:

River is the main source of water. It is a form of natural open channel which gives rise to many complex phenomenon of sciences that needs to be tackled such as the critical flow conditions, boundary shear stress and depth averaged velocity. The development of society more or less solely depends upon the flow of rivers. The rivers are major sources of many sediments and specific ingredients which are much essential for human beings. During floods, part of a river is carried by the simple main channel and rest is carried by flood plains. For such compound asymmetric channels, the flow structure becomes complicated due to momentum exchange between main channel and adjoining flood plains. Distribution of boundary shear in subsections provides us with the concept of momentum transfer between the interface of main channel and the flood plains. Experimentally, to get better data with accurate results are very complex because of the complexity of the problem. Hence, Conveyance Estimation System (CES) software has been used to tackle the complex processes to determine the shear stresses at different sections of an open channel having asymmetric flood plains on both sides of the main channel and the results are compared with the symmetric flood plains for various geometrical shapes and flow conditions. Error analysis is also performed to know the degree of accuracy of the model implemented.

Keywords: depth average velocity, non prismatic compound channel, relative flow depth , velocity distribution

Procedia PDF Downloads 104

Authors: Shogo Nagano, Yusuke Tanigaki, Hirokazu Fukuda

Abstract:

Vegetables harvested early in the morning or late in the afternoon are valued in plant production, and so the time of harvest is important. The biological functions known as circadian clocks have a significant effect on this harvest timing. The purpose of this study was to non-destructively estimate the circadian clock and so construct a method for determining a suitable harvest time. We took eight samples of green busil (Perilla frutescens var. crispa) every 4 hours, six times for 1 day and analyzed all samples at the same time. A hyperspectral camera was used to collect spectrum intensities at 141 different wavelengths (350–1050 nm). Calculation of correlations between spectrum intensity of each wavelength and harvest time suggested the suitability of the hyperspectral camera for non-destructive estimation. However, even the highest correlated wavelength had a weak correlation, so we used machine learning to raise the accuracy of estimation and constructed a machine learning model to estimate the internal time of the circadian clock. Artificial neural networks (ANN) were used for machine learning because this is an effective analysis method for large amounts of data. Using the estimation model resulted in an error between estimated and real times of 3 min. The estimations were made in less than 2 hours. Thus, we successfully demonstrated this method of non-destructively estimating internal time.

Keywords: artificial neural network (ANN), circadian clock, green busil, hyperspectral camera, non-destructive evaluation

Procedia PDF Downloads 278

Authors: Niroshan K. Walgama Wellalage, Tieling Zhang, Richard Dwight

Abstract:

State-based Markov deterioration models (SMDM) sometimes fail to find accurate transition probability matrix (TPM) values, and hence lead to invalid future condition prediction or incorrect average deterioration rates mainly due to drawbacks of existing nonlinear optimization-based algorithms and/or subjective function types used for regression analysis. Furthermore, a set of separate functions for each condition state with age cannot be directly derived by using Markov model for a given bridge element group, which however is of interest to industrial partners. This paper presents a new approach for generating Homogeneous SMDM model output, namely, the Modified Weibull approach, which consists of a set of appropriate functions to describe the percentage condition prediction of bridge elements in each state. These functions are combined with Bayesian approach and Metropolis Hasting Algorithm (MHA) based Markov Chain Monte Carlo (MCMC) simulation technique for quantifying the uncertainty in model parameter estimates. In this study, factors contributing to rail bridge deterioration were identified. The inspection data for 1,000 Australian railway bridges over 15 years were reviewed and filtered accordingly based on the real operational experience. Network level deterioration model for a typical bridge element group was developed using the proposed Modified Weibull approach. The condition state predictions obtained from this method were validated using statistical hypothesis tests with a test data set. Results show that the proposed model is able to not only predict the conditions in network-level accurately but also capture the model uncertainties with given confidence interval.

Keywords: bridge deterioration modelling, modified weibull approach, MCMC, metropolis-hasting algorithm, bayesian approach, Markov deterioration models

Procedia PDF Downloads 708

Authors: Nishal Parbhoo, John B. Dewar, Samantha Gildenhuys

Abstract:

Rotavirus is the major cause of severe gastroenteritis worldwide in children aged 5 and younger. Death rates are high particularly in developing countries. The mature rotavirus is a non-enveloped triple-layered nucleocapsid containing 11 double-stranded RNA segments. Here a global view on the sequence and structure of the three main capsid proteins, VP7, VP6, and VP2 is taken by generating a consensus sequence for each of these rotavirus proteins, for each species obtained from published data of representative rotavirus genotypes from across the world and across species. The degree of conservation between species was represented on homology models for each of the proteins. VP7 shows the highest level of variation with 14 - 45 amino acids showing conservation of less than 60%. These changes are localized to the outer surface which is exposed to antibodies alluding to a possible mechanism in evading the immune system. The middle layer, VP6 shows lower variability with only 14-32 sites having lower than 70% conservation. The inner structural layer made up of VP2 showed the lowest variability with only 1-16 sites having less than 70% conservation across species. The results correlate with proteins’ multiple structural roles. Although the nucleotide sequences vary due to an error-prone replication and lack of proofreading, the corresponding amino acid sequence of VP2, 6 and 7 remains conserved. Sequence conservation maintained for the virus results in stable protein structures, fit for function. This can be exploited in drug design, molecular studies and biotechnological applications.

Keywords: amino acid sequence conservation, capsid protein, protein structure, vaccine candidate

Procedia PDF Downloads 277

Authors: Wasantha Subasinghe

Abstract:

Crimes are one of big social problems in Sri Lanka. Crimes can be seen as simply way as an activity that against for the society or public law. There are offences in minor crimes and grave crimes including murder, rape, trafficking, robbery, excise, narcotic, kidnapping and so on. There are various forms of punishment such as bailing, fining, and prisoning to the death penalty. Death penalty contains the killing of an offender for an offense. There are 23 prison institutions in Sri Lanka including 03 closed prisoners and 20 remand prisons. There are 10 work camps, 02 open prison camps, 01 training school for youthful offenders and 02 correctional centers for youthful offenders. Capital punishment is legal in Sri Lanka as many other countries as India, Japan, Bangladesh, Iran and Iraq so on. When compared unconvicted prisoners from 2006-2010 there is an increase. It was 89190 in 2006 and it was 100191 in 2010. There were 28732 of convicted prisoners and it was 32128 in 2010. There were 165 Death sentences in 2006 and it was 96 in 2010. There are 540 individuals had been sentenced to death. The death penalty has not been implemented in Sri Lanka since 1976. Research problem: What are the feelings of prisoners as waiting for death?’ Objectives of the study were identifying prisoners’ point of view on their punishment and root causes for their offence. Case studies were conducted to identify the research problem and data were collected using formal interviews. Research area was Welikada prison. Stratified sampling method in probability samplings was used. Sample size was 20 cases from death penalty and life in prison prisoners and 20 from other convicted prisoners. Findings revealed causes and feelings them as offenders. They need if death penalty or freedom. Some of them need to convert death sentence to life imprisonment. They are physically and mentally damaged after their imprisonment. Lack of hope and as well as lack of welfare and rehabilitation programs they suffered their lives.

Keywords: death penalty, expectations, life imprisonment, rehabilitation

Procedia PDF Downloads 262

Authors: Xavier Lorang, Ahmadali Tahmasebimoradi, Chetra Mang, Sylvain Girard

Abstract:

The additive manufacturing processes (e.g. selective laser melting) allow us to produce lattice structures which have less weight, higher impact absorption capacity, and better thermal exchange property compared to the classical structures. Unfortunately, geometric imperfections (defects) in the lattice structures are by-products results of the manufacturing process. These imperfections decrease the lifetime and the strength of the lattice structures and alternate their mechanical responses. The objective of the paper is to present a simulation strategy which allows us to take into account the effect of the geometric imperfections on the mechanical response of the lattice structure. In the first part, an identification method of geometric imperfection parameters of the lattice structure based on point clouds is presented. These point clouds are based on tomography measurements. The point clouds are fed into the platform LATANA (LATtice ANAlysis) developed by IRT-SystemX to characterize the geometric imperfections. This is done by projecting the point clouds of each microbeam along the beam axis onto a 2D surface. Then, by fitting an ellipse to the 2D projections of the points, the geometric imperfections are characterized by introducing three parameters of an ellipse; semi-major/minor axes and angle of rotation. With regard to the calculated parameters of the microbeam geometric imperfections, a statistical analysis is carried out to determine a probability density law based on a statistical hypothesis. The microbeam samples are randomly drawn from the density law and are used to generate lattice structures. In the second part, a finite element model for the lattice structure with the simplified geometric imperfections (ellipse parameters) is presented. This numerical model is used to simulate the generated lattice structures. The propagation of the uncertainties of geometric imperfections is shown through the distribution of the computed mechanical responses of the lattice structures.

Keywords: additive manufacturing, finite element model, geometric imperfections, lattice structures, propagation of uncertainty

Procedia PDF Downloads 169

Authors: Nkechinyere Joy Olawuyi, Babajide Samuel Afolabi, Bola Ibitoye

Abstract:

This study collected a preprocessed dataset of chest radiographs and formulated a deep neural network model for detecting abnormalities. It also evaluated the performance of the formulated model and implemented a prototype of the formulated model. This was with the view to developing a deep neural network model to automatically classify abnormalities in chest radiographs. In order to achieve the overall purpose of this research, a large set of chest x-ray images were sourced for and collected from the CheXpert dataset, which is an online repository of annotated chest radiographs compiled by the Machine Learning Research Group, Stanford University. The chest radiographs were preprocessed into a format that can be fed into a deep neural network. The preprocessing techniques used were standardization and normalization. The classification problem was formulated as a multi-label binary classification model, which used convolutional neural network architecture to make a decision on whether an abnormality was present or not in the chest radiographs. The classification model was evaluated using specificity, sensitivity, and Area Under Curve (AUC) score as the parameter. A prototype of the classification model was implemented using Keras Open source deep learning framework in Python Programming Language. The AUC ROC curve of the model was able to classify Atelestasis, Support devices, Pleural effusion, Pneumonia, A normal CXR (no finding), Pneumothorax, and Consolidation. However, Lung opacity and Cardiomegaly had a probability of less than 0.5 and thus were classified as absent. Precision, recall, and F1 score values were 0.78; this implies that the number of False Positive and False Negative is the same, revealing some measure of label imbalance in the dataset. The study concluded that the developed model is sufficient to classify abnormalities present in chest radiographs into present or absent.

Keywords: transfer learning, convolutional neural network, radiograph, classification, multi-label

Procedia PDF Downloads 92

Authors: Carol Anne Hargreaves

Abstract:

A key issue in stock investment is how to select representative features for stock selection. The objective of this paper is to firstly determine whether an automated stock investment system, using machine learning techniques, may be used to identify a portfolio of growth stocks that are highly likely to provide returns better than the stock market index. The second objective is to identify the technical features that best characterize whether a stock’s price is likely to go up and to identify the most important factors and their contribution to predicting the likelihood of the stock price going up. Unsupervised machine learning techniques, such as cluster analysis, were applied to the stock data to identify a cluster of stocks that was likely to go up in price – portfolio 1. Next, the principal component analysis technique was used to select stocks that were rated high on component one and component two – portfolio 2. Thirdly, a supervised machine learning technique, the logistic regression method, was used to select stocks with a high probability of their price going up – portfolio 3. The predictive models were validated with metrics such as, sensitivity (recall), specificity and overall accuracy for all models. All accuracy measures were above 70%. All portfolios outperformed the market by more than eight times. The top three stocks were selected for each of the three stock portfolios and traded in the market for one month. After one month the return for each stock portfolio was computed and compared with the stock market index returns. The returns for all three stock portfolios was 23.87% for the principal component analysis stock portfolio, 11.65% for the logistic regression portfolio and 8.88% for the K-means cluster portfolio while the stock market performance was 0.38%. This study confirms that an automated stock investment system using machine learning techniques can identify top performing stock portfolios that outperform the stock market.

Keywords: machine learning, stock market trading, logistic regression, cluster analysis, factor analysis, decision trees, neural networks, automated stock investment system

Procedia PDF Downloads 136

Authors: Kuan-Jen Lai, Fang-Yi Lin, Shang-Rong Huang, Yun-Zheng Zeng, Po-Chieh Hsu, Jay Wu

Abstract:

The normalized glandular dose (DgN) estimates the energy deposition of mammography in clinical practice. The Monte Carlo simulations frequently use uniformly mixed phantom for calculating the conversion factor. However, breast tissues are not uniformly distributed, leading to errors of conversion factor estimation. This study constructed a three-layer phantom to estimated more accurate of normalized glandular dose. In this study, MCNP code (Monte Carlo N-Particles code) was used to create the geometric structure. We simulated three types of target/filter combinations (Mo/Mo, Mo/Rh, Rh/Rh), six voltages (25 ~ 35 kVp), six HVL parameters and nine breast phantom thicknesses (2 ~ 10 cm) for the three-layer mammographic phantom. The conversion factor for 25%, 50% and 75% glandularity was calculated. The error of conversion factors compared with the results of the American College of Radiology (ACR) was within 6%. For Rh/Rh, the difference was within 9%. The difference between the 50% average glandularity and the uniform phantom was 7.1% ~ -6.7% for the Mo/Mo combination, voltage of 27 kVp, half value layer of 0.34 mmAl, and breast thickness of 4 cm. According to the simulation results, the regression analysis found that the three-layer mammographic phantom at 0% ~ 100% glandularity can be used to accurately calculate the conversion factors. The difference in glandular tissue distribution leads to errors of conversion factor calculation. The three-layer mammographic phantom can provide accurate estimates of glandular dose in clinical practice.

Keywords: Monte Carlo simulation, mammography, normalized glandular dose, glandularity

Procedia PDF Downloads 171

Authors: Siphat Lim

Abstract:

This study used Autoregressive Distributed Lag (ARDL) approach to cointegration. In the long-run the general price level and exchange rate have a positively significant effect on domestic output. The estimated result further revealed that fiscal stimulus help stimulate domestic output in the long-run, but not in the short-run, while monetary expansion help to stimulate output in both short-run and long-run. The result is complied with the theory which is the macroeconomic policies, fiscal and monetary policy; help to stimulate domestic output in the long-run. The estimated result of the Vector Error Correction Model (VECM) has indicated more clearly that the consumer price index has a positive effect on output with highly statistically significant. Increasing in the general price level would increase the competitiveness among producers than increase in the output. However, the exchange rate also has a positive effect and highly significant on the gross domestic product. The exchange rate depreciation might increase export since the purchasing power of foreigners has increased. More importantly, fiscal stimulus would help stimulate the domestic output in the long-run since the coefficient of government expenditure is positive. In addition, monetary expansion would also help stimulate the output and the result is highly significant. Thus, fiscal stimulus and monetary expansionary would help stimulate the domestic output in the long-run in Cambodia.

Keywords: fiscal policy, monetary policy, ARDL, VECM

Procedia PDF Downloads 404

Authors: Mogale Sabone, Thabiso Ntlole

Abstract:

The pressure from globalization has made manufacturing organizations to move towards three major competitive arenas: quality, cost, and responsiveness. Quality is a universal value and has become a global issue. In order to survive and be able to provide customers with good products, manufacturing organizations’ supporting systems, tools, and structures it uses must grow or evolve. The majority of quality management concepts and strategies that are practiced recently are aimed at detecting and correcting problems which already exist and serve to limit losses. In agile manufacturing environment there is no room for defect and error so it needs a quality management which is proactively directed at problem prevention. This proactive quality management avoids losses by focusing on failure prevention, virtual elimination of the possibility of premature failure, mistake-proofing, and assuring consistently high quality in the definition and design of creation processes. To achieve this, a decision support tool for quality control and management is suggested. Current decision support tools/methods used by most manufacturing companies in Botswana for quality management and control are not integrated, for example they are not consistent since some tests results data is recorded manually only whilst others are recorded electronically. It is only a set of procedures not a tool. These procedures cannot offer interactive decision support. This point brings to light the aim of this research which is to develop a framework which will help manufacturing companies in Botswana build a decision support tool for quality control and management.

Keywords: decision support tool, manufacturing, quality control, quality management

Procedia PDF Downloads 548

Authors: Camille Lacan, Pierre Desmet

Abstract:

In the crowdfunding model, a campaign succeeds by collecting the funds required over a predefined duration. The success of a CF campaign depends both on the capacity to attract members of the online communities concerned, and on the community members’ involvement in online word-of-mouth recommendations. To maximize the campaign's success probability, project creators (i.e., an organization appealing for financial resources) send messages to contributors to ask them to issue word of mouth. Internet users relay information about projects through Word of Mouth which is defined as “a critical tool for facilitating information diffusion throughout online communities”. The effectiveness of these messages depends on the message framing and the time at which they are sent to contributors (i.e., at the start of the campaign or close to the deadline). This article addresses the following question: What are the effect of message framing and temporal distance on the willingness to share word of mouth? Drawing on Perspectives Theory and Construal Level Theory, this study examines the interplay between message framing (Gains vs. Losses) and temporal distance (message while the deadline is coming vs. far) on intention to share word of mouth. A between-subject experimental design is conducted to test the research model. Results show significant differences between a loss-framed message (lack of benefits if the campaign fails) associated with a short deadline (ending tomorrow) compared to a gain-framed message (benefits if the campaign succeeds) associated with a distant deadline (ending in three months). However, this effect is moderated by the anticipated regret of a campaign failure and the temporal orientation. These moderating effects contribute to specifying the boundary condition of the framing effect. Handling the message framing and the temporal distance are thus the key decisions to influence the willingness to share word of mouth.

Keywords: construal levels, crowdfunding, message framing, word of mouth

Procedia PDF Downloads 229

Authors: Amr Sadek, Abdelrahaman Al-Qahtany, Turkey Salem Al-Qahtany

Abstract:

In this study, a backpropagation artificial neural network (ANN) model has been used to predict the cooling and heating capacities of air conditioners (AC) under different conditions. Sufficiently large measurement results were obtained from the national energy-efficiency laboratories in Saudi Arabia and were used for the learning process of the ANN model. The parameters affecting the performance of the AC, including temperature, humidity level, specific heat enthalpy indoors and outdoors, and the air volume flow rate of indoor units, have been considered. These parameters were used as inputs for the ANN model, while the cooling and heating capacity values were set as the targets. A backpropagation ANN model with two hidden layers and one output layer could successfully correlate the input parameters with the targets. The characteristics of the ANN model including the input-processing, transfer, neurons-distance, topology, and training functions have been discussed. The performance of the ANN model was monitored over the training epochs and assessed using the mean squared error function. The model was then used to predict the performance of the AC under conditions that were not included in the measurement results. The optimum performance of the AC was also predicted under the different environmental conditions in Saudi Arabia. The uncertainty of the ANN model predictions has been evaluated taking into account the randomness of the data and lack of learning.

Keywords: artificial neural network, uncertainty of model predictions, efficiency of air conditioners, cooling and heating capacities

Procedia PDF Downloads 48

Authors: Azam Fazelpour, Saeed R. Dehghani, Vlastimil Masek, Yuri S. Muzychka

Abstract:

This study employs a method based on image analyses and structure information to detect accumulated ice on known structures. The icing of marine vessels and offshore structures causes significant reductions in their efficiency and creates unsafe working conditions. Image processing methods are used to measure ice loads automatically. Most image processing methods are developed based on captured image analyses. In this method, ice loads on structures are calculated by defining structure coordinates and processing captured images. A pyramidal structure is designed with nine cylindrical bars as the known structure of experimental setup. Unsymmetrical ice accumulated on the structure in a cold room represents the actual case of experiments. Camera intrinsic and extrinsic parameters are used to define structure coordinates in the image coordinate system according to the camera location and angle. The thresholding method is applied to capture images and detect iced structures in a binary image. The ice thickness of each element is calculated by combining the information from the binary image and the structure coordinate. Averaging ice diameters from different camera views obtains ice thicknesses of structure elements. Comparison between ice load measurements using this method and the actual ice loads shows positive correlations with an acceptable range of error. The method can be applied to complex structures defining structure and camera coordinates.

Keywords: camera calibration, ice detection, ice load measurements, image processing

Procedia PDF Downloads 349

Authors: Harman Jaggi, David Steinsaltz, Shripad Tuljapurkar

Abstract:

Understanding the effect of fluctuations on populations is crucial in the context of increasing habitat fragmentation, climate change, and biological invasions, among others. Migration in response to environmental disturbances enables populations to escape unfavorable conditions, benefit from new environments and thereby ride out fluctuations in variable environments. Would populations disperse if there is no uncertainty? Karlin showed in 1982 that when sub-populations experience distinct but fixed growth rates at different sites, greater mixing of populations will lower the overall growth rate relative to the most favorable site. Here we ask if and when environmental variability favors migration over no-migration. Specifically, in random environments, would a small amount of migration increase the overall long-run growth rate relative to the zero migration case? We use analysis and simulations to show how long-run growth rate changes with migration rate. Our results show that when fitness (dis)advantages fluctuate over time across sites, migration may allow populations to benefit from variability. When there is one best site with highest growth rate, the effect of migration on long-run growth rate depends on the difference in expected growth between sites, scaled by the variance of the difference. When variance is large, there is a substantial probability of an inferior site experiencing higher growth rate than its average. Thus, a high variance can compensate for a difference in average growth rates between sites. Positive correlations in growth rates across sites favor less migration. With multiple sites and large fluctuations, the length of shortest cycle (excursion) from the best site (on average) matters, and we explore the interplay between excursion length, average differences between sites and the size of fluctuations. Our findings have implications for conservation biology: even when there are superior sites in a sea of poor habitats, variability and habitat quality across space may be key to determining the importance of migration.

Keywords: migration, variable-environments, random, dispersal, fluctuations, habitat-quality

Procedia PDF Downloads 122

Authors: Davit Shahnazaryan, Diogo Gomes, Mher Safaryan

Abstract:

Many symbolic computations and manipulations required in the analysis of partial differential equations (PDE) or systems of PDEs are tedious and error-prone. These computations arise when determining conservation laws, entropies or integral identities, which are essential tools for the study of PDEs. Here, we discuss a new Mathematica package for the symbolic analysis of PDEs that automate multiple tasks, saving time and effort. Methodologies: During the research, we have used concepts of linear algebra and partial differential equations. We have been working on creating algorithms based on theoretical mathematics to find results mentioned below. Major Findings: Our package provides the following functionalities; finding symmetry group of different PDE systems, generation of polynomials invariant with respect to different symmetry groups; simplification of integral quantities by integration by parts and null Lagrangian cleaning, computing general forms of expressions by integration by parts; finding equivalent forms of an integral expression that are simpler or more symmetric form; determining necessary and sufficient conditions on the coefficients for the positivity of a given symbolic expression. Conclusion: Using this package, we can simplify integral identities, find conserved and dissipated quantities of time-dependent PDE or system of PDEs. Some examples in the theory of mean-field games and semiconductor equations are discussed.

Keywords: partial differential equations, symbolic computation, conserved and dissipated quantities, mathematica

Procedia PDF Downloads 142

Authors: Geno Paul S. Cumla, Jan Mariel M. Gentiles, M. Brenda Gajelan-Samson

Abstract:

Eutrophication is a process where in there is a surplus of nutrients present in a lake or river. Harmful cyanobacteria, hypoxia, and primarily algae, which contain toxins, grow because of the excess nutrients. Algal blooms can cause fish kills, limiting the light penetration which reduces growth of aquatic organisms, causing die-offs of plants and produce conditions that are dangerous to aquatic and human life. The main cause for eutrophication is the presence of excessive amounts of phosphorus (P) and nitrogen (N). Nitrogen is necessary for the production of the plant tissues and is usually used to synthesize proteins. Nitrate is a compound that contains nitrogen, and at elevated levels it can cause harmful effects. Excessive amounts of phosphorus, displaced through human activity, is the major cause of algae growth and as well as degraded water quality. To accomplish this study the Assessment of Soluble inorganic nitrogen (SIN), Assessment of Soluble reactive phosphate (SRP), Determination of Chlorophyll a (Chl-a) concentration, and Determination of Dominating Taxa were done. The study addresses the high probability of algal blooms in Maqueda Bay by assessing the biotic and abiotic factors of Antiao and Jiabong rivers. The data predicts the overgrowth of algae and to create awareness to prevent the event from taking place. The study assesses the adverse effects that could be prevented by understanding and controlling algae. This should predict future cases of algal blooms and allow government agencies which require data to create programs to prevent and assess these issues.

Keywords: eutrophication, chlorophyll a, nitrogen, phosphorus, red tide, Kjeldahl method, spectrophotometer, assessment of soluble inorganic nitrogen, SIN, assessment of soluble reactive phosphate, SRP

Procedia PDF Downloads 121

Authors: Haggay Etkes

Abstract:

This study provides evidence for the impact of monthly interviews conducted for the Israeli Labor Force Surveys (LFSs) on estimated flows between labor force (LF) statuses and on coefficients in fixed-effects estimations. The study uses the natural experiment of parallel interviews for the quarterly and the monthly LFSs in Israel in 2011 for demonstrating that the Labor Force Participation (LFP) rate of Jewish persons who participated in the monthly LFS increased between interviews, while in the quarterly LFS it decreased. Interestingly, the estimated impact on the LFP rate of self-reporting individuals is 2.6–3.5 percentage points while the impact on the LFP rate of individuals whose data was reported by another member of their household (a proxy), is lower and statistically insignificant. The relative increase of the LFP rate in the monthly survey is a result of a lower rate of exit from the LF and a somewhat higher rate of entry into the LF relative to these flows in the quarterly survey. These differing flows have a bearing on labor search models as the monthly survey portrays a labor market with less friction and a “steady state” LFP rate that is 5.9 percentage points higher than the quarterly survey. The study also demonstrates that monthly interviews affect a specific group (45–64 year-olds); thus the sign of coefficient of age as an explanatory variable in fixed-effects regressions on LFP is negative in the monthly survey and positive in the quarterly survey.

Keywords: measurement error, surveys, search, LFSs

Procedia PDF Downloads 247

Authors: Feng Yang

Abstract:

Machine learning (ML), especially deep learning (DL), has been extensively applied to many applications in recently years and gained great success in solving different problems, including scientific problems. However, conventional ML/DL methodologies are purely data-driven which have the limitations, such as need of ample amount of labelled training data, lack of consistency to physical principles, and lack of generalizability to new problems/domains. Recently, there is a growing consensus that ML models need to further take advantage of prior knowledge to deal with these limitations. Physics-informed machine learning, aiming at integration of physics/domain knowledge into ML, has been recognized as an emerging area of research, especially in the recent 2 to 3 years. In this work, physics-informed ML, specifically physics-informed neural network (NN), is employed and implemented to estimate the displacements at x, y, z directions in a solid mechanics problem that is controlled by equilibrium equations with boundary conditions. By incorporating the physics (i.e. the equilibrium equations) into the learning process of NN, it is showed that the NN can be trained very efficiently with a small set of labelled training data. Experiments with different settings of the NN model and the amount of labelled training data were conducted, and the results show that very high accuracy can be achieved in fulfilling the equilibrium equations as well as in predicting the displacements, e.g. in setting the overall displacement of 0.1, a root mean square error (RMSE) of 2.09 × 10−4 was achieved.

Keywords: deep learning, neural network, physics-informed machine learning, solid mechanics

Procedia PDF Downloads 126