Search results for: generational models

Authors: Matan Cohen, Maxim Shoshany

Abstract:

Climatic, topographic and geological diversity, together with frequent disturbance and recovery cycles, produce highly complex spatial patterns of trees, shrubs, dwarf shrubs and bare ground patches. Assessment of spatial and temporal variations of these life-forms patterns under climate change is of high ecological priority. Here we report on one of the first attempts to discriminate between images of three Mediterranean life-forms patterns at three densities. The development of an extensive database of orthophoto images representing these 9 pattern categories was instrumental for training and testing pre-trained and newly-trained DL models utilizing DenseNet architecture. Both models demonstrated the advantages of using Deep Learning approaches over existing spectral and spatial (pattern or texture) algorithmic methods in differentiation 9 life-form spatial mixtures categories.

Keywords: texture classification, deep learning, desert fringe ecosystems, climate change

Procedia PDF Downloads 77

Authors: Masa Popovac, Philip Fine

Abstract:

Information and Communication Technologies (ICTs) have altered our social environments, and young people in particular have immersed themselves in the digital age. Despite countless benefits, younger ICT users are being exposed to various online risks such as contact with strangers, viewing of risky content, sending or receiving sexually themed images or comments (i.e. ‘sexting’) as well as cyber bullying. Parents may not be fully aware of the online spaces their children inhabit and often struggle to implement effective mediation strategies. This quantitative study explored (i) three types of online risks (contact risks, content risks and conduct risks), (ii) cyber bullying victimization and perpetration, and (iii) parental mediation among a sample of 689 South African adolescents aged between 12-17 years. Survey data was also collected for 227 of their parents relating to their perceptions of their child’s online experiences. A comparison between adolescent behaviors and parental perceptions was examined on the three variables in the study. Findings reveal various online risk taking behaviors. In terms of contact risks, 56% of adolescents reported having contact with at least one online stranger, with many meeting these strangers in person. Content risks included exposure to harmful information such as websites promoting extreme diets or self-harm as well as inappropriate content: 84% of adolescents had seen violent content and 75% had seen sexual content online. Almost 60% of adolescents engaged in conduct risks such as sexting. Eight online victimization behaviors were examined in the study and 79% of adolescents had at least one of these negative experiences, with a third (34%) defining this experience as cyber bullying. A strong connection between victimization and perpetration was found, with 63% of adolescents being both a victim and perpetrator. Very little parental mediation of ICT use was reported. Inferential statistics revealed that parents consistently underestimated their child’s online risk taking behaviors as well as their cyber bullying victimization and perpetration. Parents also overestimated mediation strategies in the home. The generational gap in the knowledge and use of ICTs is a barrier to effective parental mediation and online safety, since many negative online experiences by adolescents go undetected and can continue for extended periods of time thereby exacerbating the potential psychological and emotional distress. The study highlights the importance of including parents in online safety efforts.

Keywords: cyber bullying, online risk behaviors, parental mediation, South Africa

Procedia PDF Downloads 464

Authors: Rutej R. Mehta, Michael A. Chappell

Abstract:

Introduction: Arterial spin labelling (ASL) is an increasingly popular perfusion MRI technique, in which arterial blood water is magnetically labelled in the neck before flowing into the brain, providing a non-invasive measure of cerebral blood flow (CBF). The accuracy of ASL CBF measurements, however, is hampered by dispersion effects; the distortion of the ASL labelled bolus during its transit through the vasculature. In spite of this, the current recommended implementation of ASL – the white paper (Alsop et al., MRM, 73.1 (2015): 102-116) – does not account for dispersion, which leads to the introduction of errors in CBF. Given that the transport time from the labelling region to the tissue – the arterial transit time (ATT) – depends on the region of the brain and the condition of the patient, it is likely that these errors will also vary with the ATT. In this study, various dispersion models are assessed in comparison with the white paper (WP) formula for CBF quantification, enabling the errors introduced by the WP to be quantified. Additionally, this study examines the relationship between the errors associated with the WP and the ATT – and how this is influenced by dispersion. Methods: Data were simulated using the standard model for pseudo-continuous ASL, along with various dispersion models, and then quantified using the formula in the WP. The ATT was varied from 0.5s-1.3s, and the errors associated with noise artefacts were computed in order to define the concept of significant error. The instantaneous slope of the error was also computed as an indicator of the sensitivity of the error with fluctuations in ATT. Finally, a regression analysis was performed to obtain the mean error against ATT. Results: An error of 20.9% was found to be comparable to that introduced by typical measurement noise. The WP formula was shown to introduce errors exceeding 20.9% for ATTs beyond 1.25s even when dispersion effects were ignored. Using a Gaussian dispersion model, a mean error of 16% was introduced by using the WP, and a dispersion threshold of σ=0.6 was determined, beyond which the error was found to increase considerably with ATT. The mean error ranged from 44.5% to 73.5% when other physiologically plausible dispersion models were implemented, and the instantaneous slope varied from 35 to 75 as dispersion levels were varied. Conclusion: It has been shown that the WP quantification formula holds only within an ATT window of 0.5 to 1.25s, and that this window gets narrower as dispersion occurs. Provided that the dispersion levels fall below the threshold evaluated in this study, however, the WP can measure CBF with reasonable accuracy if dispersion is correctly modelled by the Gaussian model. However, substantial errors were observed with other common models for dispersion with dispersion levels similar to those that have been observed in literature.

Keywords: arterial spin labelling, dispersion, MRI, perfusion

Procedia PDF Downloads 356

Authors: Lluís Bermúdez, Isabel Morillo

Abstract:

Road safety involves carrying out a determined and explicit policy to reduce accidents. In the city of Barcelona, through the Local Road Safety Plan 2013-2018, in line with the framework that has been established at the European and state level, a series of preventive, corrective and technical measures are specified, with the priority objective of reducing the number of serious injuries and fatalities. In this work, based on the data from the accidents managed by the local police during the period 2010-2019, an analysis is carried out to verify whether the measures established in the Plan to reduce the accident rate have had an effect or not and to what extent. The analysis focuses on the type of accident and the type of vehicles involved. Different count regression models have been fitted, from which it can be deduced that the number of serious and fatal victims of the accidents that have occurred in the city of Barcelona has been reduced as the measures approved by the authorities.

Keywords: accident reduction, count regression models, road safety, urban traffic

Procedia PDF Downloads 111

Authors: Ian O'Loughlin

Abstract:

Memory networks aim to integrate some of the recent successes in machine learning with a dynamic memory base that can be updated and deployed in artificial reasoning tasks. These models involve training networks to identify, update, and operate over stored elements in a large memory array in order, for example, to ably perform question and answer tasks parsing real-world and simulated discourses. This family of approaches still faces numerous challenges: the performance of these network models in simulated domains remains considerably better than in open, real-world domains, wide-context cues remain elusive in parsing words and sentences, and even moderately complex sentence structures remain problematic. This innovation, employing an array of stored and updatable ‘memory’ elements over which the system operates as it parses text input and develops responses to questions, is a compelling one for at least two reasons: first, it addresses one of the difficulties that standard machine learning techniques face, by providing a way to store a large bank of facts, offering a way forward for the kinds of long-term reasoning that, for example, recurrent neural networks trained on a corpus have difficulty performing. Second, the addition of a stored long-term memory component in artificial reasoning seems psychologically plausible; human reasoning appears replete with invocations of long-term memory, and the stored but dynamic elements in the arrays of memory networks are deeply reminiscent of the way that human memory is readily and often characterized. However, this apparent psychological plausibility is belied by a recent turn in the study of human memory in cognitive science. In recent years, the very notion that there is a stored element which enables remembering, however dynamic or reconstructive it may be, has come under deep suspicion. In the wake of constructive memory studies, amnesia and impairment studies, and studies of implicit memory—as well as following considerations from the cognitive neuroscience of memory and conceptual analyses from the philosophy of mind and cognitive science—researchers are now rejecting storage and retrieval, even in principle, and instead seeking and developing models of human memory wherein plasticity and dynamics are the rule rather than the exception. In these models, storage is entirely avoided by modeling memory using a recurrent neural network designed to fit a preconceived energy function that attains zero values only for desired memory patterns, so that these patterns are the sole stable equilibrium points in the attractor network. So although the array of long-term memory elements in memory networks seem psychologically appropriate for reasoning systems, they may actually be incurring difficulties that are theoretically analogous to those that older, storage-based models of human memory have demonstrated. The kind of emergent stability found in the attractor network models more closely fits our best understanding of human long-term memory than do the memory network arrays, despite appearances to the contrary.

Keywords: artificial reasoning, human memory, machine learning, neural networks

Procedia PDF Downloads 254

Authors: Abobaker Mohammed Qasem Farhan, Olof M. A. Saif

Abstract:

The demand for parking spaces in urban areas, particularly in developing countries, has led to a significant issue in the absence of sufficient parking spaces in crowded areas, which results in daily traffic congestion as drivers search for parking. This not only affects the appearance of the city but also has indirect impacts on the economy, society, and environment. In response to these challenges, researchers from various countries have sought technical and intelligent solutions to mitigate the problem through the development of smart parking systems. This paper aims to analyze and design three models of parking lots, with a focus on parking time and security. The study used computer software and Uppaal tools to simulate the models and determine the best among them. The results and suggestions provided in the paper aim to reduce the parking problems and improve the overall efficiency and safety of the parking process. The conclusion of the study highlights the importance of utilizing advanced technology to address the pressing issue of insufficient parking spaces in urban areas.

Keywords: preliminaries, system requirements, timed Au- tomata, Uppaal

Procedia PDF Downloads 119

Authors: Mark Jackson

Abstract:

Description: This paper is based on an theoretical analysis of the efficacy of the dominant model of policing in western jurisdictions. Those results are then compared with a similar analysis of a traditional reactive model. It is found that neither model provides for optimal delivery of services. Instead optimal service can be achieved by a synchronous hybrid model, termed the Convectory Policing approach. Methodology and Findings: For over three decades problem oriented policing (PO) has been the dominant model for western police agencies. Initially based on the work of Goldstein during the 1970s the problem oriented framework has spawned endless variants and approaches, most of which embrace a problem solving rather than a reactive approach to policing. This has included the Area Policing Concept (APC) applied in many smaller jurisdictions in the USA, the Scaled Response Policing Model (SRPM) currently under trial in Western Australia and the Proactive Pre-Response Approach (PPRA) which has also seen some success. All of these, in some way or another, are largely based on a model that eschews a traditional reactive model of policing. Convectory Policing (CP) is an alternative model which challenges the underpinning assumptions which have seen proliferation of the PO approach in the last three decades and commences by questioning the economics on which PO is based. It is argued that in essence, the PO relies on an unstated, and often unrecognised assumption that resources will be available to meet demand for policing services, while at the same time maintaining the capacity to deploy staff to develop solutions to the problems which were ultimately manifested in those same calls for service. The CP model relies on the observations from a numerous western jurisdictions to challenge the validity of that underpinning assumption, particularly in fiscally tight environment. In deploying staff to pursue and develop solutions to underpinning problems, there is clearly an opportunity cost. Those same staff cannot be allocated to alternative duties while engaged in a problem solution role. At the same time, resources in use responding to calls for service are unavailable, while committed to that role, to pursue solutions to the problems giving rise to those same calls for service. The two approaches, reactive and PO are therefore dichotomous. One cannot be optimised while the other is being pursued. Convectory Policing is a pragmatic response to the schism between the competing traditional and contemporary models. If it is not possible to serve either model with any real rigour, it becomes necessary to taper an approach to deliver specific outcomes against which success or otherwise might be measured. CP proposes that a structured roster-driven approach to calls for service, combined with the application of what is termed a resource-effect response capacity has the potential to resolve the inherent conflict between traditional and models of policing and the expectations of the community in terms of community policing based problem solving models.

Keywords: policing, reactive, proactive, models, efficacy

Procedia PDF Downloads 466

Authors: Adriana Postal, Silvio C. Sampaio, Marcio A. Villas Boas, Josué P. Castro

Abstract:

This study deals with the estimation of reference evapotranspiration (ET₀) in an agricultural context, focusing on efficient irrigation management to meet the growing interest in the sustainable management of water resources. Given the importance of water in agriculture and its scarcity in many regions, efficient use of this resource is essential to ensure food security and environmental sustainability. The methodology used involved the application of artificial intelligence techniques, specifically Multilayer Perceptron (MLP) Artificial Neural Networks (ANNs), to predict ET₀ in the state of Paraná, Brazil. The models were trained and validated with meteorological data from the Brazilian National Institute of Meteorology (INMET), together with data obtained from a producer's weather station in the western region of Paraná. Two optimizers (SGD and Adam) and different meteorological variables, such as temperature, humidity, solar radiation, and wind speed, were explored as inputs to the models. Nineteen configurations with different input variables were tested; amidst them, configuration 9, with 8 input variables, was identified as the most efficient of all. Configuration 10, with 4 input variables, was considered the most effective, considering the smallest number of variables. The main conclusions of this study show that MLP ANNs are capable of accurately estimating ET₀, providing a valuable tool for irrigation management in agriculture. Both configurations (9 and 10) showed promising performance in predicting ET₀. The validation of the models with cultivator data underlined the practical relevance of these tools and confirmed their generalization ability for different field conditions. The results of the statistical metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Coefficient of Determination (R²), showed excellent agreement between the model predictions and the observed data, with MAE as low as 0.01 mm/day and 0.03 mm/day, respectively. In addition, the models achieved an R² between 0.99 and 1, indicating a satisfactory fit to the real data. This agreement was also confirmed by the Kolmogorov-Smirnov test, which evaluates the agreement of the predictions with the statistical behavior of the real data and yields values between 0.02 and 0.04 for the producer data. In addition, the results of this study suggest that the developed technique can be applied to other locations by using specific data from these sites to further improve ET₀ predictions and thus contribute to sustainable irrigation management in different agricultural regions. The study has some limitations, such as the use of a single ANN architecture and two optimizers, the validation with data from only one producer, and the possible underestimation of the influence of seasonality and local climate variability. An irrigation management application using the most efficient models from this study is already under development. Future research can explore different ANN architectures and optimization techniques, validate models with data from multiple producers and regions, and investigate the model's response to different seasonal and climatic conditions.

Keywords: agricultural technology, neural networks in agriculture, water efficiency, water use optimization

Procedia PDF Downloads 28

Authors: Carlos Gonzales, Zaida Quiroz, Marcos Prates

Abstract:

Several spatial variables collected at the same location that share a common spatial distribution can be modeled simultaneously through a multivariate geostatistical model that takes into account the correlation between these variables and the spatial autocorrelation. The main goal of this model is to perform spatial prediction of these variables in the region of study. Here we focus on a geostatistical multivariate formulation that relies on sharing common spatial random effect terms. In particular, the first response variable can be modeled by a mean that incorporates a shared random spatial effect, while the other response variables depend on this shared spatial term, in addition to specific random spatial effects. Each spatial random effect is defined through a Gaussian process with a valid covariance function, but in order to improve the computational efficiency when the data are large, each Gaussian process is approximated to a Gaussian random Markov field (GRMF), specifically to the block nearest neighbor Gaussian process (Block-NNGP). This approach involves dividing the spatial domain into several dependent blocks under certain constraints, where the cross blocks allow capturing the spatial dependence on a large scale, while each individual block captures the spatial dependence on a smaller scale. The multivariate geostatistical model belongs to the class of Latent Gaussian Models; thus, to achieve fast Bayesian inference, it is used the integrated nested Laplace approximation (INLA) method. The good performance of the proposed model is shown through simulations and applications for massive data.

Keywords: Block-NNGP, geostatistics, gaussian process, GRMF, INLA, multivariate models.

Procedia PDF Downloads 79

Authors: Hayriye Anıl, Görkem Kar

Abstract:

In addition to being rich in renewable energy resources, Turkey is one of the countries that promise potential in geothermal energy production with its high installed power, cheapness, and sustainability. Increasing imbalance penalties become an economic burden for organizations since geothermal generation plants cannot maintain the balance of supply and demand due to the inadequacy of the production forecasts given in the day-ahead market. A better production forecast reduces the imbalance penalties of market participants and provides a better imbalance in the day ahead market. In this study, using machine learning, deep learning, and, time series methods, the total generation of the power plants belonging to Zorlu Natural Electricity Generation, which has a high installed capacity in terms of geothermal, was estimated for the first one and two weeks of March, then the imbalance penalties were calculated with these estimates and compared with the real values. These modeling operations were carried out on two datasets, the basic dataset and the dataset created by extracting new features from this dataset with the feature engineering method. According to the results, Support Vector Regression from traditional machine learning models outperformed other models and exhibited the best performance. In addition, the estimation results in the feature engineering dataset showed lower error rates than the basic dataset. It has been concluded that the estimated imbalance penalty calculated for the selected organization is lower than the actual imbalance penalty, optimum and profitable accounts.

Keywords: machine learning, deep learning, time series models, feature engineering, geothermal energy production forecasting

Procedia PDF Downloads 91

Authors: Shani Brathwaite, Deborah Villarroel-Lamb

Abstract:

Wave run-up may be defined as the time-varying position of the landward extent of the water’s edge, measured vertically from the mean water level position. The hydrodynamics of the swash zone and the accurate prediction of maximum wave run-up, play a critical role in the study of coastal engineering. The understanding of these processes is necessary for the modeling of sediment transport, beach recovery and the design and maintenance of coastal engineering structures. However, due to the complex nature of the swash zone, there remains a lack of detailed knowledge in this area. Particularly, there has been found to be insufficient consideration of bed porosity and ultimately infiltration/exfiltration processes, in the development of wave run-up models. Theoretically, there should be an inverse relationship between maximum wave run-up and beach porosity. The greater the rate of infiltration during an event, associated with a larger bed porosity, the lower the magnitude of the maximum wave run-up. Additionally, most models have been developed using data collected on North American or Australian beaches and may have limitations when used for operational forecasting in Trinidad. This paper aims to assess the influence and significance of infiltration and exfiltration processes on wave run-up magnitudes within the swash zone. It also seeks to pay particular attention to how well various empirical formulae can predict maximum run-up on contrasting beaches in Trinidad. Traditional surveying techniques will be used to collect wave run-up and cross-sectional data on various beaches. Wave data from wave gauges and wave models will be used as well as porosity measurements collected using a double ring infiltrometer. The relationship between maximum wave run-up and differing physical parameters will be investigated using correlation analyses. These physical parameters comprise wave and beach characteristics such as wave height, wave direction, period, beach slope, the magnitude of wave setup, and beach porosity. Most parameterizations to determine the maximum wave run-up are described using differing parameters and do not always have a good predictive capability. This study seeks to improve the formulation of wave run-up by using the aforementioned parameters to generate a formulation with a special focus on the influence of infiltration/exfiltration processes. This will further contribute to the improvement of the prediction of sediment transport, beach recovery and design of coastal engineering structures in Trinidad.

Keywords: beach porosity, empirical models, infiltration, swash, wave run-up

Procedia PDF Downloads 338

Authors: Kamal KC, Zhendong Yin, Dasen Li, Zhilu Wu

Abstract:

Intra-plant disease classification based on leaf images is a challenging computer vision task due to similarities in texture, color, and shape of leaves with a slight variation of leaf spot; and external environmental changes such as lighting and background noises. Deep convolutional neural network (DCNN) has proven to be an effective tool for binary classification. In this paper, two methods for binary classification of diseased plant leaves using DCNN are presented; model created from scratch and transfer learning. Our main contribution is a thorough evaluation of 4 networks created from scratch and transfer learning of 5 pre-trained models. Training and testing of these models were performed on a plant leaf images dataset belonging to 16 distinct classes, containing a total of 22,265 images from 8 different plants, consisting of a pair of healthy and diseased leaves. We introduce a deep CNN model, Optimized MobileNet. This model with depthwise separable CNN as a building block attained an average test accuracy of 99.77%. We also present a fine-tuning method by introducing the concept of a convolutional block, which is a collection of different deep neural layers. Fine-tuned models proved to be efficient in terms of accuracy and computational cost. Fine-tuned MobileNet achieved an average test accuracy of 99.89% on 8 pairs of [healthy, diseased] leaf ImageSet.

Keywords: deep convolution neural network, depthwise separable convolution, fine-grained classification, MobileNet, plant disease, transfer learning

Procedia PDF Downloads 168

Authors: K. D. Sharma, M. Parida, S. S. Jain, Anju Saini, V. K. Katiyar

Abstract:

Air quality models form one of the most important components of an urban air quality management plan. Various statistical modeling techniques (regression, multiple regression and time series analysis) have been used to predict air pollution concentrations in the urban environment. These models calculate pollution concentrations due to observed traffic, meteorological and pollution data after an appropriate relationship has been obtained empirically between these parameters. Artificial neural network (ANN) is increasingly used as an alternative tool for modeling the pollutants from vehicular traffic particularly in urban areas. In the present paper, an attempt has been made to model traffic air pollution, specifically CO concentration using neural networks. In case of CO concentration, two scenarios were considered. First, with only classified traffic volume input and the second with both classified traffic volume and meteorological variables. The results showed that CO concentration can be predicted with good accuracy using artificial neural network (ANN).

Keywords: air quality management, artificial neural network, meteorological variables, statistical modeling

Procedia PDF Downloads 505

Authors: Roy. H. A. Lindelauf

Abstract:

Cyberattacks are complex processes consisting of multiple interwoven tasks conducted by a set of agents. Interdictions and defenses against such attacks often rely on cyber kill chain (CKC) models. A CKC is a framework that tries to capture the actions taken by a cyber attacker. There exists a growing body of literature on CKCs. Most of this work either a) describes the CKC with respect to one or more specific cyberattacks or b) discusses the tools and technologies used by the attacker at each stage of the CKC. Defenders, facing scarce resources, have to decide where to allocate their resources given the CKC and partial knowledge on the tools and techniques attackers use. In this presentation CKCs are analyzed through the lens of covert projects, i.e., interrelated tasks that have to be conducted by agents (human and/or computer) with the aim of going undetected. Various aspects of covert project models have been studied abundantly in the operations research and game theory domain, think of resource-limited interdiction actions that maximally delay completion times of a weapons project for instance. This presentation has investigated both cooperative and non-cooperative game theoretic covert project models and elucidated their relation to CKC modelling. To view a CKC as a covert project each step in the CKC is broken down into tasks and there are players of which each one is capable of executing a subset of the tasks. Additionally, task inter-dependencies are represented by a schedule. Using multi-glove cooperative games it is shown how a defender can optimize the allocation of his scarce resources (what, where and how to monitor) against an attacker scheduling a CKC. This study presents and compares several cooperative game theoretic solution concepts as metrics for assigning resources to the monitoring of agents.

Keywords: cyber defense, cyber kill chain, game theory, information warfare techniques

Procedia PDF Downloads 127

Authors: Goksel Ezgi Guzey, Bihrat Onoz

Abstract:

The scarcity of streamflow gauging stations and the increasing effects of global warming cause designing water management systems to be very difficult. This study is a significant contribution to assessing regional regression models for estimating streamflow. In this study, simulated meteorological data was related to the observed streamflow data from 1971 to 2020 for 33 stream gauging stations of the Euphrates-Tigris Basin. Ordinary least squares regression was used to predict flow for 2020-2100 with the simulated meteorological data. CORDEX- EURO and CORDEX-MENA domains were used with 0.11 and 0.22 grids, respectively, to estimate climate conditions under certain climate scenarios. Twelve meteorological variables simulated by two regional climate models, RCA4 and RegCM4, were used as independent variables in the ordinary least squares regression, where the observed streamflow was the dependent variable. The variability of streamflow was then calculated with 5-6 meteorological variables and watershed characteristics such as area and height prior to the application. Of the regression analysis of 31 stream gauging stations' data, the stations were subjected to a clustering analysis, which grouped the stations in two clusters in terms of their hydrometeorological properties. Two streamflow equations were found for the two clusters of stream gauging stations for every domain and every regional climate model, which increased the efficiency of streamflow estimation by a range of 10-15% for all the models. This study underlines the importance of homogeneity of a region in estimating streamflow not only in terms of the geographical location but also in terms of the meteorological characteristics of that region.

Keywords: hydrology, streamflow estimation, climate change, hydrologic modeling, HBV, hydropower

Procedia PDF Downloads 106

Authors: Dzhavidan Zeinalov, Bugra Sen, Firangiz Aslanova

Abstract:

Most text-to-speech models cannot operate well in low-resource languages and require a great amount of high-quality training data to be considered good enough. Yet, with the improvements made in ASR systems, it is now much easier than ever to collect data for the design of custom text-to-speech models. In this work, our work on using the ASR model to collect data to build a viable text-to-speech system for one of the leading financial institutions of Azerbaijan will be outlined. NVIDIA’s implementation of the Tacotron 2 model was utilized along with the HiFiGAN vocoder. As for the training, the model was first trained with high-quality audio data collected from the Internet, then fine-tuned on the bank’s single speaker call center data. The results were then evaluated by 50 different listeners and got a mean opinion score of 4.17, displaying that our method is indeed viable. With this, we have successfully designed the first text-to-speech model in Azerbaijani and publicly shared 12 hours of audiobook data for everyone to use.

Keywords: Azerbaijani language, HiFiGAN, Tacotron 2, text-to-speech, transfer learning, whisper

Procedia PDF Downloads 21

Authors: Vivek Mahadev, Manoj Kumar, Neelu Mathur, Brahm Dutt Pandey

Abstract:

Perimeter guarding and protection of critical installations require prompt intrusion detection and assessment to take effective countermeasures. Currently, visual and electronic surveillance are the primary methods used for perimeter guarding. These methods can be costly and complicated, requiring careful planning according to the location and terrain. Moreover, these methods often struggle to detect stealthy and camouflaged insurgents. The object of the present work is to devise a surveillance technique using seismic sensors that overcomes the limitations of existing systems. The aim is to improve intrusion detection, assessment, and characterization by utilizing seismic sensors. Most of the similar systems have only two types of intrusion detection capability viz., human or vehicle. In our work we could even categorize further to identify types of intrusion activity such as walking, running, group walking, fence jumping, tunnel digging and vehicular movements. A virtual fence of 60 meters at GCNEP, Bahadurgarh, Haryana, India, was created by installing four underground geophones at a distance of 15 meters each. The signals received from these geophones are then processed to find unique seismic signatures called features. Various feature optimization and selection methodologies, such as LightGBM, Boruta, Random Forest, Logistics, Recursive Feature Elimination, Chi-2 and Pearson Ratio were used to identify the best features for training the machine learning models. The trained models were developed using algorithms such as supervised support vector machine (SVM) classifier, kNN, Decision Tree, Logistic Regression, Naïve Bayes, and Artificial Neural Networks. These models were then used to predict the category of events, employing weighted ensemble voting to analyze and combine their results. The models were trained with 1940 training events and results were evaluated with 831 test events. It was observed that using the weighted ensemble voting increased the efficiency of predictions. In this study we successfully developed and deployed the virtual fence using geophones. Since these sensors are passive, do not radiate any energy and are installed underground, it is impossible for intruders to locate and nullify them. Their flexibility, quick and easy installation, low costs, hidden deployment and unattended surveillance make such systems especially suitable for critical installations and remote facilities with difficult terrain. This work demonstrates the potential of utilizing seismic sensors for creating better perimeter guarding and protection systems using multiple machine learning models in weighted ensemble voting. In this study the virtual fence achieved an intruder detection efficiency of over 97%.

Keywords: geophone, seismic perimeter surveillance, machine learning, weighted ensemble method

Procedia PDF Downloads 64

Authors: Chiara Mansanta, Daniela Sani

Abstract:

The present paper presents the further development of the application of a standard methodology to boost innovation inside real case studies of manufacturing companies. The proposed methodology provides a viable solution for manufacturing companies that have to evaluate new business ideas. The study underlined the concept of entrepreneurship and how a manager can use it to promote innovation inside their companies. Starting from a literature study on entrepreneurship, this paper examines the role of the manager in supporting a company’s development. The empirical part of the study is based on two manufacturing companies that used the proposed methodology to favour entrepreneurship through an alternative approach. The research demonstrated the need for companies to have a structured and well-defined methodology to achieve their goals. The purpose of this article is to understand the significance of business models inside companies and explore how they affect business strategy and innovation management. The idea is to use business models to support entrepreneurs in their decision-making processes, reducing risks and avoiding errors.

Keywords: entrepreneurship, manufacturing companies, solution validation, strategic management

Procedia PDF Downloads 80

Authors: Mostafa ‘Neo’ Mohsenvand, Mohammad Rasool Izadi, Pattie Maes

Abstract:

EEG-based biometrics (user identification) has been explored on small datasets of no more than 157 subjects. Here we show that the accuracy of modern supervised methods falls rapidly as the number of users increases to a few thousand. Moreover, supervised methods require a large amount of labeled data for training which limits their applications in real-world scenarios where acquiring data for training should not take more than a few minutes. We show that using contrastive learning for pre-training, it is possible to maintain high accuracy on a dataset of 2130 subjects while only using a fraction of labels. We compare 5 different self-supervised tasks for pre-training of the encoder where our proposed method achieves the accuracy of 96.4%, improving the baseline supervised models by 22.75% and the competing self-supervised model by 3.93%. We also study the effects of the length of the signal and the number of channels on the accuracy of the user-identification models. Our results reveal that signals from temporal and frontal channels contain more identifying features compared to other channels.

Keywords: brainprint, contrastive learning, electroencephalo-gram, self-supervised learning, user identification

Procedia PDF Downloads 143

Authors: Ting Ye, Nhan Phan-Thien, Chwee Teck Lim, Lina Peng, Huixin Shi

Abstract:

In biofluid flow systems, often the flow problems of fluids of complex structures, such as the flow of red blood cells (RBCs) through complex capillary vessels, need to be considered. In this paper, we aim to apply a particle-based method, Smoothed Dissipative Particle Dynamics (SDPD), to simulate the motion and deformation of RBCs in complex micro-tubes. We first present the theoretical models, including SDPD model, RBC-fluid interaction model, RBC deformation model, RBC aggregation model, and boundary treatment model. After that, we show the verification and validation of these models, by comparing our numerical results with the theoretical, experimental and previously-published numerical results. Finally, we provide some simulation cases, such as the motion and deformation of RBCs in rectangular, cylinder, curved, bifurcated, and constricted micro-tubes, respectively.

Keywords: aggregation, deformation, red blood cell, smoothed dissipative particle dynamics

Procedia PDF Downloads 148

Authors: Md Al Masum Bhuiyan, Maria C. Mariani, Osei K. Tweneboah

Abstract:

In the present work, we develop a technique for estimating the volatility of financial time series by using stochastic differential equation. Taking the daily closing prices from developed and emergent stock markets as the basis, we argue that the incorporation of stochastic volatility into the time-varying parameter estimation significantly improves the forecasting performance via Maximum Likelihood Estimation. While using the technique, we see the long-memory behavior of data sets and one-step-ahead-predicted log-volatility with ±2 standard errors despite the variation of the observed noise from a Normal mixture distribution, because the financial data studied is not fully Gaussian. Also, the Ornstein-Uhlenbeck process followed in this work simulates well the financial time series, which aligns our estimation algorithm with large data sets due to the fact that this algorithm has good convergence properties.

Keywords: financial time series, maximum likelihood estimation, Ornstein-Uhlenbeck type models, stochastic volatility model

Procedia PDF Downloads 223

Authors: Florian Pfeffel, Valentin Nickolai, Christian Louis Kühner

Abstract:

Digitalization has disrupted the traditional workplace environment by allowing many employees to work from anywhere at any time. This trend of working from home was further accelerated due to the COVID-19 crisis, which forced companies to rethink their workplace models. While in many companies, this shift happened out of pure necessity; many employees were left more satisfied with their job due to the opportunity to work from home. This study focuses on employees’ job satisfaction in the service sector in dependence on the different work models, which are defined as a “work from home” model, the traditional “work in office” model, and a hybrid model. Using structural equation modeling (SEM), these three work models have been analyzed based on 13 influencing factors on job satisfaction that have been further summarized in the three groups “classic influencing factors”, “influencing factors changed by remote working”, and “new remote working influencing factors”. Based on the influencing factors on job satisfaction, a survey has been conducted with n = 684 employees in the service sector. Cronbach’s alpha of the individual constructs was shown to be suitable. Furthermore, the construct validity of the constructs was confirmed by face validity, content validity, convergent validity (AVE > 0.5: CR > 0.7), and discriminant validity. Additionally, confirmatory factor analysis (CFA) confirmed the model fit for the investigated sample (CMIN/DF: 2.567; CFI: 0.927; RMSEA: 0.048). The SEM-analysis has shown that the most significant influencing factor on job satisfaction is “identification with the work” with β = 0.540, followed by “Appreciation” (β = 0.151), “Compensation” (β = 0.124), “Work-Life-Balance” (β = 0.116), and “Communication and Exchange of Information” (β = 0.105). While the significance of each factor can vary depending on the work model, the SEM-analysis shows that the identification with the work is the most significant factor in all three work models and, in the case of the traditional office work model, it is the only significant influencing factor. The study shows that employees who work entirely remotely or have a hybrid work model are significantly more satisfied with their job, with a job satisfaction score of 5.0 respectively on a scale from 1 (very dissatisfied) to 7 (very satisfied), than employees do not have the option to work from home with a score of 4.6. This comes as a result of the lower identification with the work in the model without any remote working. Furthermore, the responses indicate that it is important to consider the individual preferences of each employee when it comes to the work model to achieve overall higher job satisfaction. Thus, it can be argued that companies can profit off of more motivation and higher productivity by considering the individual work model preferences, therefore, increasing the identification with the respective work.

Keywords: home-office, identification with work, job satisfaction, new work, remote work, structural equation modeling

Procedia PDF Downloads 69

Authors: José Leandro Melgar Néris, Claudinéia Brazil, Luciane Teresa Salvi, Isabel Cristina Damin

Abstract:

The impact the climate change is not recent, the main variable in the hydrological cycle is the sequence and shortage of a drought, which has a significant impact on the socioeconomic, agricultural and environmental spheres. This study aims to characterize and quantify, based on precipitation climatic projections, the rainy and dry events in the region of the Uruguay River Basin, through the Standardized Precipitation Index (SPI). The database is the image that is part of the Intercomparison of Model Models, Phase 5 (CMIP5), which provides condition prediction models, organized according to the Representative Routes of Concentration (CPR). Compared to the normal set of climates in the Uruguay River Watershed through precipitation projections, seasonal precipitation increases for all proposed scenarios, with a low climate trend. From the data of this research, the idea is that this article can be used to support research and the responsible bodies can use it as a subsidy for mitigation measures in other hydrographic basins.

Keywords: climate change, climatic model, dry events, precipitation projections

Procedia PDF Downloads 125

Authors: Zhihua Cao, Hongfei Zhao, Zhongneng Wu

Abstract:

The development of polar-based agroforestry was necessary due to the influence of the timber market environment in China, which can promote the coordinated development of forestry and agriculture, and gain remarkable ecological, economic and social benefits. The main agroforestry models of the main poplar planting area in Huaibei plain and along the Yangtze River plain were carried out. 11 typical management models of poplar were selected to sum up: pure poplar forest, poplar-rape-soybean, poplar-wheat-soybean, poplar-rape-cotton, poplar-wheat, poplar-chicken, poplar-duck, poplar-sheep, poplar-Agaricus blazei, poplar-oil peony, poplar-fish, represented by M0-M10, respectively. 12 indexes related with economic, ecological and social benefits (annual average cost, net income, ratio of output to investment, payback period of investment, land utilization ratio, utilization ratio of light energy, improvement and system stability of ecological and production environment, product richness, labor capacity, cultural quality of labor force, sustainability) were screened out to carry on the comprehensive evaluation and analysis to 11 kinds of typical agroforestry models based on analytic hierarchy process (AHP). The results showed that the economic benefit of each agroforestry model was in the order of: M8 > M6 > M9 > M7 > M5 > M10 > M4 > M1 > M2 > M3 > M0. The economic benefit of poplar-A. blazei model was the highest (332, 800 RMB / hm²), followed by poplar-duck and poplar-oil peony model (109, 820RMB /hm², 5, 7226 RMB /hm²). The order of comprehensive benefit was: M8 > M4 > M9 > M6 > M1 > M2 > M3 > M7 > M5 > M10 > M0. The economic benefit and comprehensive benefit of each agroforestry model were higher than that of pure poplar forest. The comprehensive benefit of poplar-A. blazei model was the highest, and that of poplar-wheat model ranked second, while its economic benefit was not high. Next were poplar-oil peony and poplar-duck models. It was suggested that the model of poplar-wheat should be adopted in the plain along the Yangtze River, and the whole cycle mode of poplar-grain, popalr-A. blazei, or poplar-oil peony should be adopted in Huaibei plain, northern Anhui. Furthermore, wheat, rape, and soybean are the main crops before the stand was closed; the agroforestry model of edible fungus or Chinese herbal medicine can be carried out when the stand was closed in order to maximize the comprehensive benefit. The purpose of this paper is to provide a reference for forest farmers in the selection of poplar agroforestry model in the future and to provide the basic data for the sustainable and efficient study of poplar agroforestry in Anhui province, eastern China.

Keywords: agroforestry, analytic hierarchy process (AHP), comprehensive benefit, model, poplar

Procedia PDF Downloads 144

Authors: Natacha Fery, Guilherme L. Dalledonne, Xiangyang Zheng, Cheng Tang, Roberto Mayerle

Abstract:

A Decision Support System (DSS) for supporting decision makers in the management of the Shandong Peninsula has been developed. Emphasis has been given to coastal protection, coastal cage aquaculture and harbors. The investigations were done in the framework of a joint research project funded by the German Ministry of Education and Research (BMBF) and the Chinese Academy of Sciences (CAS). In this paper, a description of the DSS, the development of its components, and results of its application are presented. The system integrates in-situ measurements, process-based models, and a database management system. Numerical models for the simulation of flow, waves, sediment transport and morphodynamics covering the entire Bohai Sea are set up based on the Delft3D modelling suite (Deltares). Calibration and validation of the models were realized based on the measurements of moored Acoustic Doppler Current Profilers (ADCP) and High Frequency (HF) radars. In order to enable cost-effective and scalable applications, a database management system was developed. It enhances information processing, data evaluation, and supports the generation of data products. Results of the application of the DSS to the management of coastal protection, coastal cage aquaculture and harbors are presented here. Model simulations covering the most severe storms observed during the last decades were carried out leading to an improved understanding of hydrodynamics and morphodynamics. Results helped in the identification of coastal stretches subjected to higher levels of energy and improved support for coastal protection measures.

Keywords: coastal protection, decision support system, in-situ measurements, numerical modelling

Procedia PDF Downloads 177

Authors: Mohd Asrul Affendi Bin Abdullah

Abstract:

Sample size calculation is especially important for zero inflated models because a large sample size is required to detect a significant effect with this model. This paper verify how to present percentage of power approximation for categorical and then extended to zero inflated models. Wald test was chosen to determine power sample size of AIDS death rate because it is frequently used due to its approachability and its natural for several major recent contribution in sample size calculation for this test. Power calculation can be conducted when covariates are used in the modeling ‘excessing zero’ data and assist categorical covariate. Analysis of AIDS death rate study is used for this paper. Aims of this study to determine the power of sample size (N = 945) categorical death rate based on parameter estimate in the simulation of the study.

Keywords: power sample size, Wald test, standardize rate, ZINBDR

Procedia PDF Downloads 423

Authors: Ajay Kumar, Shamily Jaggi

Abstract:

In developing countries like India development of modern technologies has been a key determinant in accelerating industrialization and urbanization. But in the pursuit of rapid economic growth, development is considered a top priority, while environmental protection is not given the same importance. Thus, a number of industries sited haphazardly have been established, leading to a deterioration of natural resources like water, soil and air. As a result, environmental pollution is tremendously increasing due to industrialization and mechanization that are serving to fulfill the demands of the population. With the increasing population, demand for bricks for construction work is also increasing, establishing the brick industry as a growing industry. Brick production requires two main resources; water as a source of life, and soil, as a living environment. Water and soil conservation is a critical issue in areas facing scarcity of water and soil resources. The purpose of this review paper is to provide a brief overview of the theoretical frameworks used in the analysis of the adoption and/or acceptance of soil and water conservation practices in the brick industry. Different frameworks and models have been used in the analysis of the adoption and/or acceptance of new technologies and practices; these include the technology acceptance model, motivational model, theory of reasoned action, innovation diffusion theory, theory of planned behavior, and the unified theory of acceptance and use of technology. However, every model has some limitations, such as not considering environmental/contextual and economic factors that may affect the individual’s intention to perform a behavior. The paper concludes that in comparing other models, the UTAUT seems a better model for understanding the dynamics of acceptance and adoption of water and soil conservation practices.

Keywords: brick kiln, water conservation, soil conservation, unified theory of acceptance and use of technology, technology adoption

Procedia PDF Downloads 89

Authors: Armineh Nourbakhsh, Sameena Shah, Carolyn Rose

Abstract:

Layout-aware language models have been used to create multimodal representations for documents that are in image form, achieving relatively high accuracy in document understanding tasks. However, the large number of parameters in the resulting models makes building and using them prohibitive without access to high-performing processing units with large memory capacity. We propose an alternative approach that can create efficient representations without the need for a neural visual backbone. This leads to an 80% reduction in the number of parameters compared to the smallest SOTA model, widely expanding applicability. In addition, our layout embeddings are pre-trained on spatial and visual cues alone and only fused with text embeddings in downstream tasks, which can facilitate applicability to low-resource of multi-lingual domains. Despite using 2.5% of training data, we show competitive performance on two form understanding tasks: semantic labeling and link prediction.

Keywords: layout understanding, form understanding, multimodal document understanding, bias-augmented attention

Procedia PDF Downloads 131

Authors: Yepeng Cheng, Yasuhiko Morimoto

Abstract:

Customer relationship analysis is vital for retail stores, especially for supermarkets. The point of sale (POS) systems make it possible to record the daily purchasing behaviors of customers as an identification point of sale (ID-POS) database, which can be used to analyze customer behaviors of a supermarket. The customer value is an indicator based on ID-POS database for detecting the customer loyalty of a store. In general, there are many supermarkets in a city, and other nearby competitor supermarkets significantly affect the customer value of customers of a supermarket. However, it is impossible to get detailed ID-POS databases of competitor supermarkets. This study firstly focused on the customer value and distance between a customer's home and supermarkets in a city, and then constructed the models based on logistic regression analysis to analyze correlations between distance and purchasing behaviors only from a POS database of a supermarket chain. During the modeling process, there are three primary problems existed, including the incomparable problem of customer values, the multicollinearity problem among customer value and distance data, and the number of valid partial regression coefficients. The improved customer value, Huff’s gravity model, and inverse attractiveness frequency are considered to solve these problems. This paper presents three types of models based on these three methods for loyal customer classification and competitors’ influence analysis. In numerical experiments, all types of models are useful for loyal customer classification. The type of model, including all three methods, is the most superior one for evaluating the influence of the other nearby supermarkets on customers' purchasing of a supermarket chain from the viewpoint of valid partial regression coefficients and accuracy.

Keywords: customer value, Huff's Gravity Model, POS, Retailer

Procedia PDF Downloads 108

Authors: Mais Khader, Xingjie Wei

Abstract:

This research sheds light on female entrepreneurs by providing new insights on the survival predictions of companies led by females in the UK. This study aims to build a predictive machine learning model of the survival of female-led & co-led small & medium enterprises (SMEs) in the UK over the period 2000-2020. The predictive model built utilised a combination of financial and non-financial features related to both companies and their directors to predict SMEs' survival. These features were studied in terms of their contribution to the resultant predictive model. Five machine learning models are used in the modelling: Decision tree, AdaBoost, Naïve Bayes, Logistic regression and SVM. The AdaBoost model had the highest performance of the five models, with an accuracy of 73% and an AUC of 80%. The results show high feature importance in predicting companies' survival for company size, management experience, financial performance, industry, region, and females' percentage in management.

Keywords: company survival, entrepreneurship, females, machine learning, SMEs

Procedia PDF Downloads 73