Search results for: AFT model

Authors: L. Rashidian, M. Rajabali

Abstract:

Climate changes have unpredictable and costly effects on water resources of various basins. The impact of atmospheric phenomena on human life and the environment is so significant that only knowledge of management can reduce its consequences. In this study, using LARS.WG model and down scaling of general circulation climate model HADCM-3 and according to the IPCC scenarios, including series A1b, A2 and B1, we simulated data from 2010 to 2040 in order to using them for long term forecasting of climate parameters of the Caspian Sea and its impact on sea level. Our research involves collecting data on monthly precipitation amounts, minimum and maximum temperature and daily sunshine hours, from meteorological organization for Caspian Sea coastal station such as Gorgan, Ramsar, Rasht, Anzali, Astara and Ghaemshahr since their establishment until 2010. Considering the fact that the fluctuation range of water level in the Caspian Sea has various ups and downs in different times, there is an increase in minimum and maximum temperature for all the mentioned scenarios, which will last until 2040. Overall, the amount of rainfall in cities bordering the Caspian Sea was studied based on the three scenarios, which shows an increase in the amount. However, there will be a decrease in water level of the Caspian Sea till 2040.

Keywords: IPCC, climate change, atmospheric circulation, Caspian Sea, HADCM3, sea level

Procedia PDF Downloads 243

Authors: Zahra Ramezanpanah, Joachim Carvallo, Aurelien Rodriguez

Abstract:

This paper aims to detect anomalies in streaming data using machine learning algorithms. In this regard, we designed two separate pipelines and evaluated the effectiveness of each separately. The first pipeline, based on supervised machine learning methods, consists of two phases. In the first phase, we trained several supervised models using the UNSW-NB15 data-set. We measured the efficiency of each using different performance metrics and selected the best model for the second phase. At the beginning of the second phase, we first, using Argus Server, sniffed a local area network. Several types of attacks were simulated and then sent the sniffed data to a running algorithm at short intervals. This algorithm can display the results of each packet of received data in real-time using the trained model. The second pipeline presented in this paper is based on unsupervised algorithms, in which a Temporal Graph Network (TGN) is used to monitor a local network. The TGN is trained to predict the probability of future states of the network based on its past behavior. Our contribution in this section is introducing an indicator to identify anomalies from these predicted probabilities.

Keywords: temporal graph network, anomaly detection, cyber security, IDS

Procedia PDF Downloads 103

Authors: Robert McGrath, Zara Qureshi

Abstract:

A innovative leadership philosophy that is proposed herein is distinctly more humane than most leadership approaches Concerned Leadership. The central idea to this approach is to consider the whole person that comes to work; their professional skills and talents, as well as any personal, emotional challenges that could be affecting productivity and effectiveness at work. This paper explores Concerned Leadership as an integration of the two conceptual models areas examined in this paper –(1) leaderful organizations and practices, as well as (2) organizational culture, and defines leadership in the context of Mental Health and Wellness in the workplace. Leaderful organizations calls for organizations to implement leaderful practice. Leaderful practice is when leadership responsibility and decision-making is shared across all team members and levels, versus only delegated to top management as commonly seen. A healthy culture thrives off key aspects such as acceptance, employee pride, equal opportunity, and strong company leadership. Concerned Leadership is characterized by five main components: Self-Concern, Leaderful Practice, Human Touch, Belonging, and Compassion. As scholars and practitioners conceptualize leadership in practice, the present model seeks to uphold the dignity of each organizational member, thereby having the potential to transform workplaces and support all members.

Keywords: leadership, mental health, reflective practice, organizational culture

Procedia PDF Downloads 81

Authors: Zheng Zhihao

Abstract:

Recently, Knowledge Graph Framework Network (KGCN) has developed powerful capabilities in knowledge representation and reasoning tasks. However, traditional KGCN often uses a fixed weight mechanism when aggregating information, failing to make full use of rich structural information, resulting in a certain expression ability of node representation, and easily causing over-smoothing problems. In order to solve these challenges, the paper proposes an new graph neural network model called KGCN-STAR (Knowledge Graph Convolutional Network with Structural Adaptive Receptive Fields). This model dynamically adjusts the perception of each node by introducing a structural adaptive receptive field. wild range, and a subgraph aggregator is designed to capture local structural information more effectively. Experimental results show that KGCN-STAR shows significant performance improvement on multiple knowledge graph data sets, especially showing considerable capabilities in the task of representation learning of complex structures.

Keywords: knowledge graph, graph neural networks, structural adaptive receptive fields, information aggregation

Procedia PDF Downloads 34

Authors: Eric Pla Erra, Mariana Jimenez Martinez

Abstract:

While aggregated loads at a community level tend to be easier to predict, individual household load forecasting present more challenges with higher volatility and uncertainty. Furthermore, the drastic changes that our behavior patterns have suffered due to the COVID-19 pandemic have modified our daily electrical consumption curves and, therefore, further complicated the forecasting methods used to predict short-term electric load. Load forecasting is vital for the smooth and optimized planning and operation of our electric grids, but it also plays a crucial role for individual domestic consumers that rely on a HEMS (Home Energy Management Systems) to optimize their energy usage through self-generation, storage, or smart appliances management. An accurate forecasting leads to higher energy savings and overall energy efficiency of the household when paired with a proper HEMS. In order to study how COVID-19 has affected the accuracy of forecasting methods, an evaluation of the performance of a state-of-the-art LGBM (Light Gradient Boosting Model) will be conducted during the transition between pre-pandemic and lockdowns periods, considering day-ahead electric load forecasting. LGBM improves the capabilities of standard Decision Tree models in both speed and reduction of memory consumption, but it still offers a high accuracy. Even though LGBM has complex non-linear modelling capabilities, it has proven to be a competitive method under challenging forecasting scenarios such as short series, heterogeneous series, or data patterns with minimal prior knowledge. An adaptation of the LGBM model – called “resilient LGBM” – will be also tested, incorporating a concept drift detection technique for time series analysis, with the purpose to evaluate its capabilities to improve the model’s accuracy during extreme events such as COVID-19 lockdowns. The results for the LGBM and resilient LGBM will be compared using standard RMSE (Root Mean Squared Error) as the main performance metric. The models’ performance will be evaluated over a set of real households’ hourly electricity consumption data measured before and during the COVID-19 pandemic. All households are located in the city of Barcelona, Spain, and present different consumption profiles. This study is carried out under the ComMit-20 project, financed by AGAUR (Agència de Gestiód’AjutsUniversitaris), which aims to determine the short and long-term impacts of the COVID-19 pandemic on building energy consumption, incrementing the resilience of electrical systems through the use of tools such as HEMS and artificial intelligence.

Keywords: concept drift, forecasting, home energy management system (HEMS), light gradient boosting model (LGBM)

Procedia PDF Downloads 105

Authors: Khaled Ayfi

Abstract:

In industrial oxygen systems at high temperature and high pressure, contamination by solid particles is one of the principal causes of ignition hazards. Indeed, gas can sweep away particles, generated by corrosion inside the pipes or during maintenance operations (welding residues, careless disassembly, etc.) and produce accumulations at places where the gas velocity decrease. Moreover, in such an environment rich in oxygen (oxidant), particles are highly reactive and can ignite system walls more actively and at higher temperatures. Oxidation based thermal effects are responsible for mechanical properties lost, leading to the destruction of the pressure equipment wall. To deal with this problem, a numerical analysis is done regarding a sample representative of a wall subjected to pressure and temperature. The validation and analysis are done comparing the numerical simulations results to experimental measurements. More precisely, in this work, we propose a numerical model that describes the thermomechanical behavior of thin metal disks under pressure and subjected to laser heating. This model takes into account the geometric and material nonlinearity and has been validated by the comparison of simulation results with experimental measurements.

Keywords: ignition, oxygen, numerical simulation, thermomechanical behavior

Procedia PDF Downloads 105

Authors: Gholba Niranjan Dilip, Anil Kumar

Abstract:

Conduct of accurate reliable mapping of oil palm plantations and census of individual palm trees is a huge challenge. This study addresses this challenge and developed an optimized solution implemented deep learning techniques on remote sensing data. The oil palm is a very important tropical crop. To improve its productivity and land management, it is imperative to have accurate census over large areas. Since, manual census is costly and prone to approximations, a methodology for automated census using panchromatic images from Cartosat-2, SkySat and World View-3 satellites is demonstrated. It is selected two different study sites in Indonesia. The customized set of training data and ground-truth data are created for this study from Cartosat-2 images. The pre-trained model of Single Shot MultiBox Detector (SSD) Lite MobileNet V2 Convolutional Neural Network (CNN) from the TensorFlow Object Detection API is subjected to transfer learning on this customized dataset. The SSD model is able to generate the bounding boxes for each oil palm and also do the counting of palms with good accuracy on the panchromatic images. The detection yielded an F-Score of 83.16 % on seven different images. The detections are buffered and dissolved to generate polygons demarcating the boundaries of the oil palm plantations. This provided the area under the plantations and also gave maps of their location, thereby completing the automated census, with a fairly high accuracy (≈100%). The trained CNN was found competent enough to detect oil palm crowns from images obtained from multiple satellite sensors and of varying temporal vintage. It helped to estimate the increase in oil palm plantations from 2014 to 2021 in the study area. The study proved that high-resolution panchromatic satellite image can successfully be used to undertake census of oil palm plantations using CNNs.

Keywords: object detection, oil palm tree census, panchromatic images, single shot multibox detector

Procedia PDF Downloads 160

Authors: Alexander Goncharenko, Andrey Denisov, Sergey Alyamkin, Evgeny Terentev

Abstract:

The neural network quantization is highly desired procedure to perform before running neural networks on mobile devices. Quantization without fine-tuning leads to accuracy drop of the model, whereas commonly used training with quantization is done on the full set of the labeled data and therefore is both time- and resource-consuming. Real life applications require simplification and acceleration of quantization procedure that will maintain accuracy of full-precision neural network, especially for modern mobile neural network architectures like Mobilenet-v1, MobileNet-v2 and MNAS. Here we present a method to significantly optimize training with quantization procedure by introducing the trained scale factors for discretization thresholds that are separate for each filter. Using the proposed technique, we quantize the modern mobile architectures of neural networks with the set of train data of only ∼ 10% of the total ImageNet 2012 sample. Such reduction of train dataset size and small number of trainable parameters allow to fine-tune the network for several hours while maintaining the high accuracy of quantized model (accuracy drop was less than 0.5%). Ready-for-use models and code are available in the GitHub repository.

Keywords: distillation, machine learning, neural networks, quantization

Procedia PDF Downloads 325

Authors: Jia-Lien Hsu, Chiung-Wen Chang

Abstract:

Research articles may follow a simple and succinct structure of organizational patterns, called move. For example, considering extended abstracts, we observe that an extended abstract usually consists of five moves, including Background, Aim, Method, Results, and Conclusion. As another example, when publishing articles in PubMed, authors are encouraged to provide a structured abstract, which is an abstract with distinct and labeled sections (e.g., Introduction, Methods, Results, Discussions) for rapid comprehension. This paper introduces a method for computational analysis of move structures (i.e., Background-Purpose-Method-Result-Conclusion) in abstracts and introductions of research documents, instead of manually time-consuming and labor-intensive analysis process. In our approach, sentences in a given abstract and introduction are automatically analyzed and labeled with a specific move (i.e., B-P-M-R-C in this paper) to reveal various rhetorical status. As a result, it is expected that the automatic analytical tool for move structures will facilitate non-native speakers or novice writers to be aware of appropriate move structures and internalize relevant knowledge to improve their writing. In this paper, we propose a Bayesian approach to determine move tags for research articles. The approach consists of two phases, training phase and testing phase. In the training phase, we build a Bayesian model based on a couple of given initial patterns and the corpus, a subset of CiteSeerX. In the beginning, the priori probability of Bayesian model solely relies on initial patterns. Subsequently, with respect to the corpus, we process each document one by one: extract features, determine tags, and update the Bayesian model iteratively. In the testing phase, we compare our results with tags which are manually assigned by the experts. In our experiments, the promising accuracy of the proposed approach reaches 56%.

Keywords: academic English writing, assisted writing, move tag analysis, Bayesian approach

Procedia PDF Downloads 330

Authors: Khatesiree Sripoothon, Usanee Sengpanich, Rattana Sittioum

Abstract:

The objective of this study is to examine the factors influencing consumer purchasing decisions about healthy food. This model consists of two latent variables: Consumer Perception relating to NCDs and Consumer Perceived Product Value. The study was conducted in the northern provinces of Thailand, which are popular with tourists and have received support from the government for health tourism. A survey was used as the data collection method, and the questionnaire was applied to 385 tourists. An accidental sampling method was used to identify the sample. The statistics of frequency, percentage, mean, and structural equation model were used to analyze the data obtained. Additionally, all factors had a significant positive influence on healthy food purchasing decisions (p<0.01) and were predictive of healthy food purchasing decisions at 46.20 (R2=0.462). Also, these findings seem to underline a supposition that consumer perceptions of NCDs and perceived product value are key variables that strengthens the competitive effects of a healthy-friendly business entrepreneur. Moreover, reduce the country's public health costs for treating patients with the disease of NCDs in Thailand.

Keywords: healthy food, perceived product value, perception of non-communicable diseases, purchasing decisions

Procedia PDF Downloads 161

Authors: Nwachukwu M. A., Nwachukwu J. I., Anyanwu J., Emeka U., Okorondu J., Acholonu C.

Abstract:

Actualization of the 17 sustainable development goals (SDGs) conceived by the United Nations in 2015 is a global challenge that may not be feasible in sub-Saharan Africa by the year 2030, except universities play a committed role. This is because; there is a need to educate the people about the concepts of sustainability and sustainable development in the region to make the desired change. Here is a sensitization paper with a model of intervention and curricular planning to allow advancement in understanding and knowledge of SDGs. This Model Center for Sustainability Studies (MCSS) will enable partnerships with institutions in Africa and in advanced nations, thereby creating a global network for sustainability studies not found in sub-Saharan Africa. MCSS will train and certify public servants, government agencies, policymakers, entrepreneurs and personnel from organizations, and students on aspects of the SDGs and sustainability science. There is a need to add sustainability knowledge into environmental education and make environmental education a compulsory course in higher institutions and a secondary school certificate exam subject in sub-Saharan Africa. MCSS has 11 training modules that can be replicated anywhere in the world.

Keywords: sustainability, higher institutions, training, SDGs, collaboration, sub-Saharan Africa

Procedia PDF Downloads 99

Authors: Zafar Mahmood, Surayya Naz, Nazir Shah Khattak

Abstract:

The degree of freedom relates to our capability to observe or model the energy paths within the system. Higher the number of energy paths being modeled leaves to us a higher degree of freedom, but increasing the time and modeling complexity rendering it useless for today’s world’s need for minimum time to market. Since the number of residuals that can be uniquely isolated are dependent on the number of independent outputs of the system, increasing the number of sensors required. The examples of discrete position sensors that may be used to form an array include limit switches, Hall effect sensors, optical sensors, magnetic sensors, etc. Their mechanical design can usually be tailored to fit in the transitional path of an STME in a variety of mechanical configurations. The case studies into multi-sensor system were carried out and actual data from sensors is used to test this generic framework. It is being investigated, how the proper modeling of limit switches as timing sensors, could lead to unified and neutral residual space while keeping the implementation cost reasonably low.

Keywords: low-cost limit sensors, fault diagnostics, Single Throw Mechanical Equipment (STME), parameter estimation, parity-space

Procedia PDF Downloads 617

Authors: Nouha Ben Salah, Mohamed Ali Omri

Abstract:

This paper treats the association between disclosure level and the cost of equity capital in Tunisian’slisted firms. This relation is tested by using two models. The first is used for testing this relation directly by regressing firm specific estimates of cost of equity capital on market beta, firm size and a measure of disclosure level. The second model is used for testing this relation by introducing information asymmetry as mediator variable. This model is suggested by Baron and Kenny (1986) to demonstrate the role of mediator variable in general. Based on a sample of 21 non-financial Tunisian’s listed firms over a period from 2000 to 2004, the results prove that greater disclosure is associated with a lower cost of equity capital. However, the results of indirect relationship indicate a significant positive association between the level of voluntary disclosure and information asymmetry and a significant negative association between information asymmetry and cost of equity capital in contradiction with our previsions. Perhaps this result is due to the biases of measure of information asymmetry.

Keywords: cost of equity capital, voluntary disclosure, information asymmetry, and Tunisian’s listed non-financial firms

Procedia PDF Downloads 517

Authors: M. Kosandi, V. Susanti, N. I. Subono, E. Kartini

Abstract:

This paper discusses the risk of glorification trap in combating human trafficking, as it is shown in the case of Indonesia. Based on a research on Indonesian combat against trafficking in 2017-2018, this paper shows the tendency of misinterpretation and misapplication of the Indonesian anti-trafficking law into misusing the law for glorification, to create an image of certain extent of achievement in combating human trafficking. The objective of this paper is to explain the persistent occurrence of human trafficking crimes despite the significant progress of anti-trafficking efforts of Indonesian government. The research was conducted in 2017-2018 by qualitative approach through observation, depth interviews, discourse analysis, and document study, applying the three-dimensional model for analyzing human trafficking in the source country. This paper argues that the drive for glorification of achievement in the combat against trafficking has trapped Indonesian government in the loop of misinterpretation, misapplication, and misuse of the anti-trafficking law. In return, the so-called crime against humanity remains high and tends to increase in Indonesia.

Keywords: human trafficking, anti-trafficking policy, transnational crime, source country, glorification trap

Procedia PDF Downloads 167

Authors: Shereif. H. Mahmoud

Abstract:

Egypt has limited water resources, and it will be under water stress by the year 2030. Therefore, Egypt should consider natural and non-conventional water resources to overcome such a problem. Rain harvesting is one solution. This Paper presents a geographic information system (GIS) methodology - based on decision support system (DSS) that uses remote sensing data, filed survey, and GIS to identify potential RWH areas. The input into the DSS includes a map of rainfall surplus, slope, potential runoff coefficient (PRC), land cover/use, soil texture. In addition, the outputs are map showing potential sites for RWH. Identifying suitable RWH sites implemented in the ArcGIS model environment using the model builder of ArcGIS 10.1. Based on Analytical hierarchy process (AHP) analysis taking into account five layers, the spatial extents of RWH suitability areas identified using Multi-Criteria Evaluation (MCE). The suitability model generated a suitability map for RWH with four suitability classes, i.e. Excellent, Moderate, Poor, and unsuitable. The spatial distribution of the suitability map showed that the excellent suitable areas for RWH concentrated in the northern part of Egypt. According to their averages, 3.24% of the total area have excellent and good suitability for RWH, while 45.04 % and 51.48 % of the total area are moderate and unsuitable suitability, respectively. The majority of the areas with excellent suitability have slopes between 2 and 8% and with an intensively cultivated area. The major soil type in the excellent suitable area is loam and the rainfall range from 100 up to 200 mm. Validation of the used technique depends on comparing existing RWH structures locations with the generated suitability map using proximity analysis tool of ArcGIS 10.1. The result shows that most of exiting RWH structures categorized as successful.

Keywords: rainwater harvesting (RWH), geographic information system (GIS), analytical hierarchy process (AHP), multi-criteria evaluation (MCE), decision support system (DSS)

Procedia PDF Downloads 360

Authors: Desiree Vandenberghe, Johan Albrecht

Abstract:

Non-communicable diseases (NCDs) are responsible for the vast majority of deaths in the European Union (EU) and represent a large share of total health care spending. A future increase in this health and financial burden is likely to be driven by population ageing, lifestyle changes and technological advances in medicine. Without adequate prevention measures, this burden can severely threaten population health and economic development. To tackle this challenge, a correct assessment of the current burden of NCDs is required, as well as a projection of potential increases of this burden. The contribution of this paper is to offer perspective on the evolution of the NCD burden towards the future and to give an indication of the potential of prevention policy. A Non-Homogenous, Semi-Markov model for the EU was constructed, which allowed for a projection of the cost burden for the four main NCDs (cancer, cardiovascular disease, chronic respiratory disease and diabetes mellitus) towards 2030 and 2050. This simulation is done based on multiple baseline scenarios that vary in demand and supply factors such as health status, population structure, and technological advances. Finally, in order to assess the potential of preventive measures to curb the cost explosion of NCDs, a simulation is executed which includes increased efforts for preventive health care measures. According to the Markov model, by 2030 and 2050, total costs (direct and indirect costs) in the EU could increase by 30.1% and 44.1% respectively, compared to 2015 levels. An ambitious prevention policy framework for NCDs will be required if the EU wants to meet this challenge of rising costs. To conclude, significant cost increases due to Non-Communicable Diseases are likely to occur due to demographic and lifestyle changes. Nevertheless, an ambitious prevention program throughout the EU can aid in making this cost burden manageable for future generations.

Keywords: non-communicable diseases, preventive health care, health policy, Markov model, scenario analysis

Procedia PDF Downloads 140

Authors: Feng-Sheng Wang, Chao-Ting Cheng

Abstract:

Developing a drug from conception to launch is costly and time-consuming. Computer-aided methods can reduce research costs and accelerate the development process during the early drug discovery and development stages. This study developed a fuzzy multi-objective hierarchical optimization framework for identifying potential anticancer targets in a metabolic model. First, RNA-seq expression data of colorectal cancer samples and their healthy counterparts were used to reconstruct tissue-specific genome-scale metabolic models. The aim of the optimization framework was to identify anticancer targets that lead to cancer cell death and evaluate metabolic flux perturbations in normal cells that have been caused by cancer treatment. Four objectives were established in the optimization framework to evaluate the mortality of cancer cells for treatment and to minimize side effects causing toxicity-induced tumorigenesis on normal cells and smaller metabolic perturbations. Through fuzzy set theory, a multiobjective optimization problem was converted into a trilevel maximizing decision-making (MDM) problem. The applied nested hybrid differential evolution was applied to solve the trilevel MDM problem using two nutrient media to identify anticancer targets in the genome-scale metabolic model of colorectal cancer, respectively. Using Dulbecco’s Modified Eagle Medium (DMEM), the computational results reveal that the identified anticancer targets were mostly involved in cholesterol biosynthesis, pyrimidine and purine metabolisms, glycerophospholipid biosynthetic pathway and sphingolipid pathway. However, using Ham’s medium, the genes involved in cholesterol biosynthesis were unidentifiable. A comparison of the uptake reactions for the DMEM and Ham’s medium revealed that no cholesterol uptake reaction was included in DMEM. Two additional media, i.e., a cholesterol uptake reaction was included in DMEM and excluded in HAM, were respectively used to investigate the relationship of tumor cell growth with nutrient components and anticancer target genes. The genes involved in the cholesterol biosynthesis were also revealed to be determinable if a cholesterol uptake reaction was not induced when the cells were in the culture medium. However, the genes involved in cholesterol biosynthesis became unidentifiable if such a reaction was induced.

Keywords: Cancer metabolism, genome-scale metabolic model, constraint-based model, multilevel optimization, fuzzy optimization, hybrid differential evolution

Procedia PDF Downloads 80

Authors: Nishank Raisinghani

Abstract:

Efficiently predicting drug response remains a challenge in the realm of drug discovery. To address this issue, we propose four model architectures that combine graphical representation with varying positions of multiheaded self-attention mechanisms. By leveraging two types of multi-omics data, transcriptomics and genomics, we create a comprehensive representation of target cells and enable drug response prediction in precision medicine. A majority of our architectures utilize multiple transformer models, one with a graph attention mechanism and the other with a multiheaded self-attention mechanism, to generate latent representations of both drug and omics data, respectively. Our model architectures apply an attention mechanism to both drug and multiomics data, with the goal of procuring more comprehensive latent representations. The latent representations are then concatenated and input into a fully connected network to predict the IC-50 score, a measure of cell drug response. We experiment with all four of these architectures and extract results from all of them. Our study greatly contributes to the future of drug discovery and precision medicine by looking to optimize the time and accuracy of drug response prediction.

Keywords: drug discovery, transformers, graph neural networks, multiomics

Procedia PDF Downloads 154

Authors: Sipho Mbatha, Anne Mastament-Mason

Abstract:

South African manufacturing global competitiveness was ranked 22nd (out of 38 countries), dropped to 24th in 2013 and is expected to drop further to 25th by 2018. These impacts negatively on the industrialisation project of South Africa. For industrialization to be achieved through labour intensive industries like the Apparel Manufacturing Industries of South Africa (AMISA), South Africa needs to identify and respond to factors negatively impacting on the development of competitive advantage This paper applied factor conditions from Porter’s Diamond Model (1990) to understand the various challenges facing the AMISA. Factor conditions highlighted in Porter’s model are grouped into two groups namely, basic and advance factors. Two AMISA associations representing over 10 000 employees were interviewed. The largest Clothing, Textiles and Leather (CTL) apparel retail group was also interviewed with a government department implementing the industrialisation policy were interviewed The paper points out that while AMISA have basic factor conditions necessary for competitive advantage in the clothing and textiles industries, Advance factor coordination has proven to be a challenging task for the AMISA, Higher Education Institutions (HEIs) and government. Poor infrastructural maintenance has contributed to high manufacturing costs and poor quick response as a result of lack of advanced technologies. The use of Porter’s Factor Conditions as a tool to analyse the sector’s competitive advantage challenges and opportunities has increased knowledge regarding factors that limit the AMISA’s competitiveness. It is therefore argued that other studies on Porter’s Diamond model factors like Demand conditions, Firm strategy, structure and rivalry and Related and supporting industries can be used to analyse the situation of the AMISA for the purposes of improving competitive advantage.

Keywords: compliance rule, apparel manufacturing industry, factor conditions, advance skills and South African industrial policy

Procedia PDF Downloads 362

Authors: Smitha Haridasan, Zelalem Demissie, Atri Dutta, Ajita Rattani

Abstract:

Aided by the wind, all it takes is one ember and a few minutes to create a wildfire. Wildfires are growing in frequency and size due to climate change. Wildfires and its consequences are one of the major environmental concerns. Every year, millions of hectares of forests are destroyed over the world, causing mass destruction and human casualties. Thus early detection of wildfire becomes a critical component to mitigate this threat. Many computer vision-based techniques have been proposed for the early detection of forest fire using video surveillance. Several computer vision-based methods have been proposed to predict and detect forest fires at various spectrums, namely, RGB, HSV, and YCbCr. The aim of this paper is to propose a multi-spectral deep learning model that combines information from different spectrums at intermediate layers for accurate fire detection. A heterogeneous dataset assembled from publicly available datasets is used for model training and evaluation in this study. The experimental results show that multi-spectral deep learning models could obtain an improvement of about 4.68 % over those based on a single spectrum for fire detection.

Keywords: deep learning, forest fire detection, multi-spectral learning, natural hazard detection

Procedia PDF Downloads 241

Authors: Juan Sebastián Hernández

Abstract:

The Nevado del Ruíz volcano, located between the limits of the Departments of Caldas and Tolima in Colombia, presented an unstable behaviour in the course of the year 2020, this volcanic activity led to secondary effects on the crust, which is why the prediction of deformations becomes the task of geoscientists. In the course of this article, the use of tropospheric variables such as evapotranspiration, UV aerosol index, carbon monoxide, nitrogen dioxide, methane, surface temperature, among others, is used to train a set of neural networks that can predict the behaviour of the resulting phase of an unrolled interferogram with the DInSAR technique, whose main objective is to identify and characterise the behaviour of the crust based on the environmental conditions. For this purpose, variables were collected, a generalised linear model was created, and a set of neural networks was created. After the training of the network, validation was carried out with the test data, giving an MSE of 0.17598 and an associated r-squared of approximately 0.88454. The resulting model provided a dataset with good thematic accuracy, reflecting the behaviour of the volcano in 2020, given a set of environmental characteristics.

Keywords: crustal deformation, Tropomi, neural networks (ANN), volcanic activity, DInSAR

Procedia PDF Downloads 103

Authors: Nicolas Lopez-Villalobos, Dorian J. Garrick, Hugh T. Blair

Abstract:

Many dairy goat farmers of New Zealand milk herds of mixed breed does. Simultaneous evaluation of sires and does across breed is required to select the best animals for breeding on a common basis. Across-breed estimated breeding values (EBV) and estimated producing values for 208-day lactation yields of milk (MY), fat (FY), protein (PY) and somatic cell score (SCS; LOG2(SCC) of Saanen, Nubian, Alpine, Toggenburg and crossbred dairy goats from 75 herds were estimated using a test day model. Evaluations were based on 248,734 herd-test records representing 125,374 lactations from 65,514 does sired by 930 sires over 9 generations. Averages of MY, FY and PY were 642 kg, 21.6 kg and 19.8 kg, respectively. Average SCC and SCS were 936,518 cells/ml milk and 9.12. Pure-bred Saanen does out-produced other breeds in MY, FY and PY. Average EBV for MY, FY and PY compared to a Saanen base were Nubian -98 kg, 0.1 kg and -1.2 kg; Alpine -64 kg, -1.0 kg and -1.7 kg; and Toggenburg -42 kg, -1.0 kg and -0.5 kg. First-cross heterosis estimates were 29 kg MY, 1.1 kg FY and 1.2 kg PY. Average EBV for SCS compared to a Saanen base were Nubian 0.041, Alpine -0.083 and Toggenburg 0.094. Heterosis for SCS was 0.03. Breeding values are combined with respective economic values to calculate an economic index used for ranking sires and does to reflect farm profit.

Keywords: breed effects, dairy goats, milk traits, test-day model

Procedia PDF Downloads 330

Authors: Odete Paiva, Cláudia Seabra, José Luís Abrantes, Fernanda Cravidão

Abstract:

There is a clear ‘cult of authenticity’, at least in modern Western society. So, there is a need to analyze the tourist perception of authenticity, bearing in mind the destination, its attractions, motivations, cultural distance, and contact with other tourists. Our study seeks to investigate the relationship among cultural values, image, sense of place, perception of authenticity and behavior intentions at World Heritage Historic Centers. From a theoretical perspective, few researches focus on the impact of cultural values, image and sense of place on authenticity and intentions behavior in tourists. The intention of this study is to help close this gap. A survey was applied to collect data from tourists visiting two World Heritage Historic Centers – Guimarães in Portugal and Cordoba in Spain. Data was analyzed in order to establish a structural equation model (SEM). Discussion centers on the implications of model to theory and managerial development of tourism strategies. Recommendations for destinations managers and promoters and tourist organizations administrators are addressed.

Keywords: authenticity perception, behavior intentions, cultural tourism, cultural values, world heritage historic centers

Procedia PDF Downloads 316

Authors: S. Nandagopalan, N. Pradeep

Abstract:

The aim of this paper is to propose a general framework for storing, analyzing, and extracting knowledge from two-dimensional echocardiographic images, color Doppler images, non-medical images, and general data sets. A number of high performance data mining algorithms have been used to carry out this task. Our framework encompasses four layers namely physical storage, object identification, knowledge discovery, user level. Techniques such as active contour model to identify the cardiac chambers, pixel classification to segment the color Doppler echo image, universal model for image retrieval, Bayesian method for classification, parallel algorithms for image segmentation, etc., were employed. Using the feature vector database that have been efficiently constructed, one can perform various data mining tasks like clustering, classification, etc. with efficient algorithms along with image mining given a query image. All these facilities are included in the framework that is supported by state-of-the-art user interface (UI). The algorithms were tested with actual patient data and Coral image database and the results show that their performance is better than the results reported already.

Keywords: active contour, Bayesian, echocardiographic image, feature vector

Procedia PDF Downloads 445

Authors: Sarah E. Nelson, Casey Weiner, Alexander Sigmon, Jun Hua, Haris I. Sair, Jose I. Suarez, Robert D. Stevens

Abstract:

Graph-theoretical information from structural connectomes indicated significant connectivity changes and improved acute prognostication in a Random Forest (RF) model in aneurysmal subarachnoid hemorrhage (aSAH), which can lead to significant morbidity and mortality and has traditionally been fraught by poor methods to predict outcome. This study’s hypothesis was that structural connectivity changes occur in canonical brain networks of acute aSAH patients, and that these changes are associated with functional outcome at six months. In a prospective cohort of patients admitted to a single institution for management of acute aSAH, patients underwent diffusion tensor imaging (DTI) as part of a multimodal MRI scan. A weighted undirected structural connectome was created of each patient’s images using Constant Solid Angle (CSA) tractography, with 176 regions of interest (ROIs) defined by the Johns Hopkins Eve atlas. ROIs were sorted into four networks: Default Mode Network, Executive Control Network, Salience Network, and Whole Brain. The resulting nodes and edges were characterized using graph-theoretic features, including Node Strength (NS), Betweenness Centrality (BC), Network Degree (ND), and Connectedness (C). Clinical (including demographics and World Federation of Neurologic Surgeons scale) and graph features were used separately and in combination to train RF and Logistic Regression classifiers to predict two outcomes: dichotomized modified Rankin Score (mRS) at discharge and at six months after discharge (favorable outcome mRS 0-2, unfavorable outcome mRS 3-6). A total of 56 aSAH patients underwent DTI a median (IQR) of 7 (IQR=8.5) days after admission. The best performing model (RF) combining clinical and DTI graph features had a mean Area Under the Receiver Operator Characteristic Curve (AUROC) of 0.88 ± 0.00 and Area Under the Precision Recall Curve (AUPRC) of 0.95 ± 0.00 over 500 trials. The combined model performed better than the clinical model alone (AUROC 0.81 ± 0.01, AUPRC 0.91 ± 0.00). The highest-ranked graph features for prediction were NS, BC, and ND. These results indicate reorganization of the connectome early after aSAH. The performance of clinical prognostic models was increased significantly by the inclusion of DTI-derived graph connectivity metrics. This methodology could significantly improve prognostication of aSAH.

Keywords: connectomics, diffusion tensor imaging, graph theory, machine learning, subarachnoid hemorrhage

Procedia PDF Downloads 189

Authors: Ranjit Kumar Sahu, Ramakar Jha

Abstract:

Urban Water Management is the practice of managing freshwater, waste water, and storm water as components of a basin-wide management plan. It builds on existing water supply and sanitation considerations within an urban settlement by incorporating urban water management within the scope of the entire river basin. The pervasive problems generated by urban development have prompted, in the present work, to study the spatial extent of urbanization in Golden Triangle of Odisha connecting the cities Bhubaneswar (20.2700° N, 85.8400° E), Puri (19.8106° N, 85.8314° E) and Konark (19.9000° N, 86.1200° E)., and patterns of periodic changes in urban development (systematic/random) in order to develop future plans for (i) urbanization promotion areas, and (ii) urbanization control areas. Remote Sensing, using USGS (U.S. Geological Survey) Landsat8 maps, supervised classification of the Urban Sprawl has been done for during 1980 - 2014, specifically after 2000. This Work presents the following: (i) Time series analysis of Hydrological data (ground water and rainfall), (ii) Application of SWMM (Storm Water Management Model) and other soft computing techniques for Urban Water Management, and (iii) Uncertainty analysis of model parameters (Urban Sprawl and correlation analysis). The outcome of the study shows drastic growth results in urbanization and depletion of ground water levels in the area that has been discussed briefly. Other relative outcomes like declining trend of rainfall and rise of sand mining in local vicinity has been also discussed. Research on this kind of work will (i) improve water supply and consumption efficiency (ii) Upgrade drinking water quality and waste water treatment (iii) Increase economic efficiency of services to sustain operations and investments for water, waste water, and storm water management, and (iv) engage communities to reflect their needs and knowledge for water management.

Keywords: Storm Water Management Model (SWMM), uncertainty analysis, urban sprawl, land use change

Procedia PDF Downloads 425

Authors: Negar Sartipzadeh

Abstract:

Energy-conscious design, which coordinates with the Earth ecological systems during its life cycle, has the least negative impact on the environment with the least waste of resources. Due to the increasing in world population as well as the consumption of fossil fuels that cause the production of greenhouse gasses and environmental pollution, mankind is looking for renewable and also sustainable energies. The Iranian native construction is a clear evidence of energy-aware designing. Our predecessors were forced to rely on the natural resources and sustainable energies as well as environmental issues which have been being considered in the recent world. One of these endless energies is wind energy. Iranian traditional architecture foundations is a appropriate model in solving the environmental crisis and the contemporary energy. What will come in this paper is an effort to recognition and introduction of the unique characteristics of the Iranian architecture in the application of aerodynamic and hydraulic energies derived from the wind, which are the most common and major type of using sustainable energies in the traditional architecture of Iran. Therefore, the recent research attempts to offer a hybrid system suggestions for application in new constructions designing in a region such as Nashtifan, which has potential through reviewing windmills and how they deal with sustainable energy sources, as a model of Iranian native construction.

Keywords: renewable energy, sustainable building, windmill, Iranian architecture

Procedia PDF Downloads 422

Authors: Massimo Miccoli, Luca Marangoni, Alberto Aniello Scaringi, Alessandro Marceddu, Alessandro Amicone

Abstract:

The susceptibility of deep neural networks (DNNs) to adversarial manipulations is a recognized challenge within the computer vision domain. Adversarial examples, crafted by adding subtle yet malicious alterations to benign images, exploit this vulnerability. Various defense strategies have been proposed to safeguard DNNs against such attacks, stemming from diverse research hypotheses. Building upon prior work, our approach involves the utilization of autoencoder models. Autoencoders, a type of neural network, are trained to learn representations of training data and reconstruct inputs from these representations, typically minimizing reconstruction errors like mean squared error (MSE). Our autoencoder was trained on a dataset of benign examples; learning features specific to them. Consequently, when presented with significantly perturbed adversarial examples, the autoencoder exhibited high reconstruction errors. The architecture of the autoencoder was tailored to the dimensions of the images under evaluation. We considered various image sizes, constructing models differently for 256x256 and 512x512 images. Moreover, the choice of the computer vision model is crucial, as most adversarial attacks are designed with specific AI structures in mind. To mitigate this, we proposed a method to replace image-specific dimensions with a structure independent of both dimensions and neural network models, thereby enhancing robustness. Our multi-modal autoencoder reconstructs the spectral representation of images across the red-green-blue (RGB) color channels. To validate our approach, we conducted experiments using diverse datasets and subjected them to adversarial attacks using models such as ResNet50 and ViT_L_16 from the torch vision library. The autoencoder extracted features used in a classification model, resulting in an MSE (RGB) of 0.014, a classification accuracy of 97.33%, and a precision of 99%.

Keywords: adversarial attacks, malicious images detector, binary classifier, multimodal transformer autoencoder

Procedia PDF Downloads 113

Authors: Julia Zimmerman, Gaurav Savant

Abstract:

This study aims to evaluate the efficacy of the U.S. Army Corp of Engineers’ River Analysis System (HEC-RAS) application to modeling the hydraulics of estuaries. HEC-RAS has been broadly used for a variety of riverine applications. However, it has not been widely applied to the study of circulation in estuaries. This report details the model development and validation of a combined 1D/2D unsteady flow hydraulic model using HEC-RAS for estuaries and they are associated with tidally influenced rivers. Two estuaries, Galveston Bay and Delaware Bay, were used as case studies. Galveston Bay, a bar-built, vertically mixed estuary, was modeled for the 2005 calendar year. Delaware Bay, a drowned river valley estuary, was modeled from October 22, 2019, to November 5, 2019. Water surface elevation was used to validate both models by comparing simulation results to NOAA’s Center for Operational Oceanographic Products and Services (CO-OPS) gauge data. Simulations were run using the Diffusion Wave Equations (DW), the Shallow Water Equations, Eulerian-Lagrangian Method (SWE-ELM), and the Shallow Water Equations Eulerian Method (SWE-EM) and compared for both accuracy and computational resources required. In general, the Diffusion Wave Equations results were found to be comparable to the two Shallow Water equations sets while requiring less computational power. The 1D/2D combined approach was valid for study areas within the 2D flow area, with the 1D flow serving mainly as an inflow boundary condition. Within the Delaware Bay estuary, the HEC-RAS DW model ran in 22 minutes and had an average R² value of 0.94 within the 2-D mesh. The Galveston Bay HEC-RAS DW ran in 6 hours and 47 minutes and had an average R² value of 0.83 within the 2-D mesh. The longer run time and lower R² for Galveston Bay can be attributed to the increased length of the time frame modeled and the greater complexity of the estuarine system. The models did not accurately capture tidal effects within the 1D flow area.

Keywords: Delaware bay, estuarine hydraulics, Galveston bay, HEC-RAS, one-dimensional modeling, two-dimensional modeling

Procedia PDF Downloads 199

Authors: Rim Chang Hyon, Song Hak Jin, Song Kwang Hyok, Jong Ki Hun

Abstract:

The calibration of the articulated arm coordinate measuring machine (AACMM) is key to improving calibration accuracy and saving calibration time. To reduce the time consumed for calibration, we should choose the proper calibration gauges and develop a reasonable calibration method. In addition, we should get the exact optimal solution by accurately removing the rough errors within the experimental data. In this paper, we present a calibration method of the portable coordinate measuring arm (PCMA) using the 1.2m long bar guage with cone-holes. First, we determine the locations of the bar gauge and establish an optimal objective function for identifying the structural parameter errors. Next, we make a mathematical model of the calibration algorithm and present a new mathematical method to remove the rough errors within calibration data. Finally, we find the optimal solution to identify the kinematic parameter errors by using Levenberg-Marquardt algorithm. The experimental results show that our calibration method is very effective in saving the calibration time and improving the calibration accuracy.

Keywords: AACMM, kinematic model, parameter identify, measurement accuracy, calibration

Procedia PDF Downloads 83