Search results for: back propagation neural network model

Authors: Sh. Wafa, M. Alanoud, S. Liyakathunisa

Abstract:

Online booking for a trip or hotel has become an indispensable traveling tool today, people tend to be more interested in selecting air flight travel as their first choice when going for a long trip. People's shopping behavior has greatly changed by the advent of social network. Traditional ticket booking methods are considered as outdated with the advancement in tools and technology. Web based booking framework is an 'absolute necessity to have' for any visit or movement business that is investing heaps of energy noting telephone calls, sending messages or considering employing more staff. In this paper, we propose a process model for online trip reservation for our designed web application. Our proposed system will be highly beneficial and helps in reduction in time and cost for customers.

Keywords: trip, hotel, reservation, process model, time, cost, web app

Procedia PDF Downloads 214

Authors: Chris Neale, Jenny Roe, Peter Aspinall, Sara Tilley, Steve Cinderby, Panos Mavros, Richard Coyne, Neil Thin, Catharine Ward Thompson

Abstract:

This research directly assesses older people’s neural activation in response to walking through a changing urban environment, as measured by electroencephalography (EEG). As the global urban population is predicted to grow, there is a need to understand the role that the urban environment may play on the health of its older inhabitants. There is a large body of evidence suggesting green space has a beneficial restorative effect, but this effect remains largely understudied in both older people and by using a neuroimaging assessment. For this study, participants aged 65 years and over were required to walk between a busy urban built environment and a green urban environment, in a counterbalanced design, wearing an Emotiv EEG headset to record real-time neural responses to place. Here we report on the outputs for these responses derived from both the proprietary Affectiv Suite software, which creates emotional parameters with a real time value assigned to them, as well as the raw EEG output focusing on alpha and beta changes, associated with changes in relaxation and attention respectively. Each walk lasted around fifteen minutes and was undertaken at the natural walking pace of the participant. The two walking environments were compared using a form of high dimensional correlated component regression (CCR) on difference data between the urban busy and urban green spaces. For the Emotiv parameters, results showed that levels of ‘engagement’ increased in the urban green space (with a subsequent decrease in the urban busy built space) whereas levels of ‘excitement’ increased in the urban busy environment (with a subsequent decrease in the urban green space). In the raw data, low beta (13 – 19 Hz) increased in the urban busy space with a subsequent decrease shown in the green space, similar to the pattern shown with the ‘excitement’ result. Alpha activity (9 – 13 Hz) shows a correlation with low beta, but not with dependent change in the regression model. This suggests that alpha is acting as a suppressor variable. These results suggest that there are neural signatures associated with the experience of urban spaces which may reflect the age of the cohort or the spatiality of the settings themselves. These are shown both in the outputs of the proprietary software as well as the raw EEG output. Built busy urban spaces appear to induce neural activity associated with vigilance and low level stress, while this effect is ameliorated in the urban green space, potentially suggesting a beneficial effect on attentional capacity in urban green space in this participant group. The interaction between low beta and alpha requires further investigation, in particular the role of alpha in this relationship.

Keywords: ageing, EEG, green space, urban space

Procedia PDF Downloads 224

Authors: Thrivikraman Aswathi, S. Advaith

Abstract:

As there can be cases where the use of real data is somehow limited, such as when it is hard to get access to a large volume of real data, we need to go for synthetic data generation. This produces high-quality synthetic data while maintaining the statistical properties of a specific dataset. In the present work, a generative adversarial network (GAN) is trained to produce multivariate time series (MTS) data since the MTS is now being gathered more often in various real-world systems. Furthermore, the GAN-generated MTS data is fed into a transformer-based deep learning architecture that carries out the data categorization into predefined classes. Further, the model is evaluated across various distinct domains by generating corresponding MTS data.

Keywords: GAN, transformer, classification, multivariate time series

Procedia PDF Downloads 130

Authors: P. Makendran, M. Pragadeesh, N. Narash, N. Manikandan, A. Rajasri, V. Sanal Kumar

Abstract:

The growing severity of government-obligatory emissions legislation has required continuous improvement in catalysts performance and the associated reactor systems. IC engines emit a lot of harmful gases into the atmosphere. These gases are toxic in nature and a catalytic converter is used to convert these toxic gases into less harmful gases. The catalytic converter converts these gases by Oxidation and reduction reaction. Stoichiometric engines usually use the three-way catalyst (TWC) for simultaneously destroying all of the emissions. CO and NO react to form CO2 and N2 over one catalyst, and the remaining CO and HC are oxidized in a subsequent one. Literature review reveals that typically precious metals are used as a catalyst. The actual reactor is composed of a washcoated honeycomb-style substrate, with the catalyst being contained in the washcoat. The main disadvantage of a catalytic converter is that it exerts a back pressure to the exhaust gases while entering into them. The objective of this paper is to optimize the back pressure developed by the catalytic converter through geometric optimization of catalystic converter. This can be achieved by designing a catalyst with a optimum cone angle and a more surface area of the catalyst substrate. Additionally, the arrangement of the pores in the catalyst substrate can be changed. The numerical studies have been carried out using k-omega turbulence model with varying inlet angle of the catalytic converter and the length of the catalyst substrate. We observed that the geometry optimization is a meaningful objective for the lucrative design optimization of a catalytic converter for industrial applications.

Keywords: catalytic converter, emission control, reactor systems, substrate for emission control

Procedia PDF Downloads 906

Authors: Seppo Kittilä, Sonja Dahlgren

Abstract:

In this paper, a (morpho)phonological typology of proximal and distal demonstratives is proposed. Only the most basic proximal (‘this’) and distal (‘that’) forms have been considered, potential more fine-grained distinctions based on proximity are not relevant to our discussion, nor are the other functions the discussed demonstratives may have. The sample comprises 82 languages that represent the linguistic diversity of the world’s languages, although the study is not based on a systematic sample. Four different major types are distinguished; (1) Vowel type: front vs. back; high vs. low vowel (2) Consonant type: front-back consonants (3) Additional element –type (4) Varia. The proposed types can further be subdivided according to whether the attested difference concern only, e.g., vowels, or whether there are also other changes. For example, the first type comprises both languages such as Betta Kurumba, where only the vowel changes (i ‘this’, a ‘that’) and languages like Alyawarra (nhinha vs. nhaka), where there are also other changes. In the second type, demonstratives are distinguished based on whether the consonants are front or back; typically front consonants (e.g., labial and dental) appear on proximal demonstratives and back consonants on distal demonstratives (such as velar or uvular consonants). An example is provided by Bunaq, where bari marks ‘this’ and baqi ‘that’. In the third type, distal demonstratives typically have an additional element, making it longer in form than the proximal one (e.g., Òko òne ‘this’, ònébé ‘that’), but the type also comprises languages where the distal demonstrative is simply phonologically longer (e.g., Ngalakan nu-gaʔye vs. nu-gunʔbiri). Finally, the last type comprises cases that do not fit into the three other types, but a number of strategies are used by the languages of this group. The two first types can be explained by iconicity; front or high phonemes appear on the proximal demonstratives, while back/low phonemes are related to distal demonstratives. This means that proximal demonstratives are pronounced at the front and/or high part of the oral cavity, while distal demonstratives are pronounced lower and more back, which reflects the proximal/distal nature of their referents in the physical world. The first type is clearly the most common in our data (40/82 languages), which suggests a clear association with iconicity. Our findings support earlier findings that proximal and distal demonstratives have an iconic phonemic manifestation. For example, it has been argued that /i/ is related to smallness (small distance). Consonants, however, have not been considered before, or no systematic correspondences have been discovered. The third type, in turn, can be explained by markedness; the distal element is more marked than the proximal demonstrative. Moreover, iconicity is relevant also here: some languages clearly use less linguistic substance for referring to entities close to the speaker, which is manifested in the longer (morpho)phonological form of the distal demonstratives. The fourth type contains different kinds of cases, and systematic generalizations are hard to make.

Keywords: demonstratives, iconicity, language typology, phonology

Procedia PDF Downloads 154

Authors: Aurelien Boutin

Abstract:

In optical systems such as Variable Optical Delay Lines, where a collimated beam has to go back and forth, corner cubes are used in order to keep the reflected beam parallel to the incoming beam. However, the reflected beam can be laterally shifted, which will lead to losses. In this paper, we report on a patented optical design that allows keeping the reflected beam with the exact same position and direction whatever the displacement of the corner cube leading to zero losses. After explaining how the optical design works and theoretically allows to compensate for any defects in the translation of the corner cube, we will present the results of experimental comparisons between a standard layout (i.e., only corner cubes) and our optical layout. To compare both optical layouts, we used a fiber-to-fiber coupling setup. It consists of a couple of lights from one fiber to the other, thanks to two lenses. The ensemble [fiber+lense] is fixed and called a collimator so that the light is coupled from one collimator to another. Each collimator was precisely made in order to have a precise working distance. In the experiment, we measured and compared the Insertion Losses (IL) variations between both collimators with the distance between them (i.e., natural Gaussian beam coupling losses) and between both collimators in the different optical layouts tested, with the same optical length propagation. We will show that the IL variations of our setup are less than 0.05dB with respect to the IL variations of collimators alone.

Keywords: free-space optics, variable optical delay lines, optical cavity, auto-alignment

Procedia PDF Downloads 100

Authors: Kritiyaporn Kunsook

Abstract:

Machine learning algorithms (MLAs) such us artificial neural networks (ANNs), decision tree, support vector machines (SVMs), Naïve Bayes, and ensemble classifier by voting are powerful data driven methods that are relatively less widely used in the mapping of technique of system, and thus have not been comparatively evaluated together thoroughly in this field. The performances of a series of MLAs, ANNs, decision tree, SVMs, Naïve Bayes, and ensemble classifier by voting in technique of hydroponic systems prospectively modeling are compared based on the accuracy of each model. Classification of hydroponic systems only covers the test samples from vegetables grown with Nutrient film technique (NFT) and Deep flow technique (DFT). The feature, which are the characteristics of vegetables compose harvesting height width, temperature, require light and color. The results indicate that the classification performance of the ANNs is 98%, decision tree is 98%, SVMs is 97.33%, Naïve Bayes is 96.67%, and ensemble classifier by voting is 98.96% algorithm respectively.

Keywords: artificial neural networks, decision tree, support vector machines, naïve Bayes, ensemble classifier by voting

Procedia PDF Downloads 372

Authors: Asma Ben Yaghlane, Mohamed Naceur Azaiez, Mehdi Mrad

Abstract:

We investigate the large scale of networks in the context of network survivability under attack. We use appropriate techniques to evaluate and the attacker-based- and the defender-based-network survivability. The attacker is unaware of the operated links by the defender. Each attacked link has some pre-specified probability to be disconnected. The defender choice is so that to maximize the chance of successfully sending the flow to the destination node. The attacker however will select the cut-set with the highest chance to be disabled in order to partition the network. Moreover, we extend the problem to the case of selecting the best p paths to operate by the defender and the best k cut-sets to target by the attacker, for arbitrary integers p,k > 1. We investigate some variations of the problem and suggest polynomial-time solutions.

Keywords: defense/attack strategies, large scale, networks, partitioning a network

Procedia PDF Downloads 283

Authors: Vishal H. Shukla, Sanjay B. Deshmukh

Abstract:

Traffic engineering, at its center, is the ability of moving traffic approximately so that traffic from a congested link is moved onto the unused capacity on another link. Traffic Engineering ensures the best possible use of the resources. Now to support traffic engineering in the today’s network, Multiprotocol Label Switching (MPLS) is being used which is very helpful for reliable packets delivery in an ongoing internet services. Here a topology is been implemented on GNS3 to focus on the analysis of the communication take place from one site to other through the ISP. The comparison is made between the IP network & MPLS network based on Bandwidth & Jitter which are one of the performance parameters using JPERF simulator.

Keywords: GNS3, JPERF, MPLS, traffic engineering, VMware

Procedia PDF Downloads 487

Authors: Yalçın Bozkurt

Abstract:

Prediction models were developed for accurate prediction of bodyweight (BW) by using Digital Images of beef cattle body dimensions by Artificial Neural Networks (ANN). For this purpose, the animal data were collected at a private slaughter house and the digital images and the weights of each live animal were taken just before they were slaughtered and the body dimensions such as digital wither height (DJWH), digital body length (DJBL), digital body depth (DJBD), digital hip width (DJHW), digital hip height (DJHH) and digital pin bone length (DJPL) were determined from the images, using the data with 1069 observations for each traits. Then, prediction models were developed by ANN. Digital body measurements were analysed by ANN for body prediction and R2 values of DJBL, DJWH, DJHW, DJBD, DJHH and DJPL were approximately 94.32, 91.31, 80.70, 83.61, 89.45 and 70.56 % respectively. It can be concluded that in management situations where BW cannot be measured it can be predicted accurately by measuring DJBL and DJWH alone or both DJBD and even DJHH and different models may be needed to predict BW in different feeding and environmental conditions and breeds

Keywords: artificial neural networks, bodyweight, cattle, digital body measurements

Procedia PDF Downloads 372

Authors: P. W. Tsai, W. L. Hong, C. W. Chen, C. Y. Chen

Abstract:

In this paper, we present a neural network (NN) based approach represent a nonlinear Tagagi-Sugeno (T-S) system. A linear differential inclusion (LDI) state-space representation is utilized to deal with the NN models. Taking advantage of the LDI representation, the stability conditions and controller design are derived for a class of nonlinear structural systems. Moreover, the concept of utilizing the Parallel Particle Swarm Optimization (PPSO) algorithm to solve the common P matrix under the stability criteria is given in this paper.

Keywords: Lyapunov stability, parallel particle swarm optimization, linear differential inclusion, artificial intelligence

Procedia PDF Downloads 656

Authors: Mohamed Eldessouki, Ebraheem Shady, Yasser Gowayed

Abstract:

Carbon nanotubes (CNTs) are known for having high elastic properties with high surface area that promote them as good candidates for reinforcing polymeric matrices. In composite materials, CNTs lack chemical bonding with the surrounding matrix which decreases the possibility of better stress transfer between the components. In this work, a chemical treatment for activating the surface of the multi-wall carbon nanotubes (MWCNT) was applied and the effect of this functionalization on the elastic properties of the epoxy nanocomposites was studied. Functional amino-groups were added to the surface of the CNTs and it was evaluated to be about 34% of the total weight of the CNTs. Elastic modulus was found to increase by about 40% of the neat epoxy resin at CNTs’ weight fraction of 0.5%. The elastic modulus was found to decrease after reaching a certain concentration of CNTs which was found to be 1% wt. The scanning electron microscopic pictures showed the effect of the CNTs on the crack propagation through the sample by forming stress concentrated spots at the nanocomposite samples.

Keywords: carbon nanotubes functionalization, crack propagation, elastic modulus, epoxy nanocomposites

Procedia PDF Downloads 405

Authors: Zeta Dooly, Aidan Duane

Abstract:

The collaborative European funded research and development landscape provides prime environmental conditions for multi-disciplinary teams to learn and enhance their knowledge beyond the capability of training and learning within their own organisation cocoons. Whilst the emergence of the academic entrepreneur has changed the focus of educational institutions to that of quasi-businesses, the training and professional development of lecturers and academic staff are often not formalised to the same level as industry. This research focuses on industry and academic collaborative research funded by the European Commission. The impact of research is scalable if an optimum research network is created and managed effectively. This paper investigates network embeddedness, the nature of relationships, links, and nodes within a research network, and the enhancement of the network’s knowledge. The contribution of this paper extends our understanding of establishing and maintaining effective collaborative research networks. The effects of network embeddedness are recognized in the literature as pertinent to innovation and the economy. Network theory literature claims that networks are essential to innovative clusters such as Silicon valley and innovation in high tech industries. This research provides evidence to support the impact collaborative research has on the disparate individuals toward their innovative contributions to their organisations and their own professional development. This study adopts a qualitative approach and uncovers some of the challenges of multi-disciplinary research through case study insights. The contribution of this paper recommends the establishment of scaffolding to accommodate cooperation in research networks, role appointment, and addressing contextual complexities early to avoid problem cultivation. Furthermore, it suggests recommendations in relation to network formation, intra-network challenges in relation to open data, competition, friendships, and competency enhancement. The network capability is enhanced by the adoption of the relevant theories; network theory, open innovation, and social exchange, with the understanding that the network structure has an impact on innovation and social exchange in research networks. The research concludes that there is an opportunity to deepen our understanding of the impact of network reuse and network hoping that provides scaffolding for the network members to enhance and build upon their knowledge using a progressive approach.

Keywords: research networks, competency building, network theory, case study

Procedia PDF Downloads 126

Authors: Xun Sun, Le Li, Ce Mo, Lei Mo, Ruiming Wang, Guosheng Ding

Abstract:

Cognitive control plays a central role in information processing, which is comprised of various components including response suppression and inhibitory control. Response suppression is considered to inhibit the irrelevant response during the cognitive process; while inhibitory control to inhibit the irrelevant stimulus in the process of cognition. Both of them undertake distinct functions for the cognitive control, so as to enhance the performances in behavior. Among numerous factors on cognitive control, bilingual experience is a substantial and indispensible factor. It has been reported that bilingual experience can influence the neural activity of cognitive control as whole. However, it still remains unknown how the neural influences specifically present on the components of cognitive control imposed by bilingualism. In order to explore the further issue, the study applied fMRI, used anti-saccade paradigm and compared the cerebral activations between high and low proficient Chinese-English bilinguals. Meanwhile, the study provided experimental evidence for the brain plasticity of language, and offered necessary bases on the interplay between language and cognitive control. The results showed that response suppression recruited the middle frontal gyrus (MFG) in low proficient Chinese-English bilinguals, but the inferior patrietal lobe in high proficient Chinese-English bilinguals. Inhibitory control engaged the superior temporal gyrus (STG) and middle temporal gyrus (MTG) in low proficient Chinese-English bilinguals, yet the right insula cortex was more active in high proficient Chinese-English bilinguals during the process. These findings illustrate insights that bilingual experience has neural influences on different components of cognitive control. Compared with low proficient bilinguals, high proficient bilinguals turn to activate advanced neural areas for the processing of cognitive control. In addition, with the acquisition and accumulation of language, language experience takes effect on the brain plasticity and changes the neural basis of cognitive control.

Keywords: bilingual experience, cognitive control, inhibition control, response suppression

Procedia PDF Downloads 483

Authors: S. Lecheb, A. Nour, A. Chellil, H. Mechakra, A. Zeggane, H. Kebir

Abstract:

In this work, repaired crack in 6061-T6 aluminum plate with composite patches presented, firstly we determine the displacement, strain, and stress, also the first six mode shape of the plate, secondly we took the same model adding central crack initiation, which is located in the center of the plate, its size vary from 20 mm to 60 mm and we compare the first results with second. Thirdly, we repair various cracks with the composite patch (carbon/epoxy) and for (2 layers, 4 layers). Finally, the comparison of stress, strain, displacement and six first natural frequencies between un-cracked specimen, crack propagation and composite patch repair.

Keywords: composite patch repair, crack growth, aluminum alloy plate, stress

Procedia PDF Downloads 599

Authors: Fazıl Gökgöz, Fahrettin Filiz

Abstract:

Electricity prices have sophisticated features such as high volatility, nonlinearity and high frequency that make forecasting quite difficult. Electricity price has a volatile and non-random character so that, it is possible to identify the patterns based on the historical data. Intelligent decision-making requires accurate price forecasting for market traders, retailers, and generation companies. So far, many shallow-ANN (artificial neural networks) models have been published in the literature and showed adequate forecasting results. During the last years, neural networks with many hidden layers, which are referred to as DNN (deep neural networks) have been using in the machine learning community. The goal of this study is to investigate electricity price forecasting performance of the shallow-ANN and DNN models for the Turkish day-ahead electricity market. The forecasting accuracy of the models has been evaluated with publicly available data from the Turkish day-ahead electricity market. Both shallow-ANN and DNN approach would give successful result in forecasting problems. Historical load, price and weather temperature data are used as the input variables for the models. The data set includes power consumption measurements gathered between January 2016 and December 2017 with one-hour resolution. In this regard, forecasting studies have been carried out comparatively with shallow-ANN and DNN models for Turkish electricity markets in the related time period. The main contribution of this study is the investigation of different shallow-ANN and DNN models in the field of electricity price forecast. All models are compared regarding their MAE (Mean Absolute Error) and MSE (Mean Square) results. DNN models give better forecasting performance compare to shallow-ANN. Best five MAE results for DNN models are 0.346, 0.372, 0.392, 0,402 and 0.409.

Keywords: deep learning, artificial neural networks, energy price forecasting, turkey

Procedia PDF Downloads 292

Authors: Leo Latasch, Mario Di Gennaro

Abstract:

Introduction: The goal of this project was to increase the resilience of the population as well as rescue organizations to make both quality and time-related improvements in handling crises. A helper network was created for this purpose. Methods: Social questions regarding the structure and purpose of helper networks were considered - specifically with regard to helper motivation, the level of commitment and collaboration between populations and agencies. The exchange of information, the coordinated use of volunteers, and the distribution of available resources will be ensured through defined communication and cooperation routines. Helper smartphones will also be used provide a picture of the situation on the ground. Results: The helper network was established and deployed based on the RESIBES information technology system. It consists of a service platform, a web portal and a smartphone app. The service platform is the central element for collaboration between the various rescue organizations, as well as for persons, associations, and companies from the population offering voluntary aid. The platform was used for: Registering helpers and resources and then requesting and assigning it in case of a disaster. These services allow the population's resources to be organized. The service platform also allows for a secure data exchange between services and external systems. Conclusions: The social and technical work priorities have allowed us to cover a full cycle of advance structural work, gaining an overview, damage management, evaluation, and feedback on experiences. This cycle allows experiences gained while handling the crisis to feed back into the cycle and improve preparations and management strategies.

Keywords: coordination, disaster, resilience, volunteers

Procedia PDF Downloads 142

Authors: Andrew Kurochkin, Kostiantyn Bokhan

Abstract:

In this study, a new meme dataset of ~650K meme instances was created, a technology of meme generation based on the state of the art deep learning technique - GPT-2 model was researched, a comparative analysis of machine-generated memes and human-created was conducted. We justified that Amazon Mechanical Turk workers can be used for the approximate estimating of users' behavior in a social network, more precisely to measure engagement. It was shown that generated memes cause the same engagement as human memes that produced low engagement in the social network (historically). Thus, generated memes are less engaging than random memes created by humans.

Keywords: content generation, computational social science, memes generation, Reddit, social networks, social media interaction

Procedia PDF Downloads 138

Authors: Vidhi Jain

Abstract:

Communication between speakers and non-speakers of American Sign Language (ASL) can be problematic, inconvenient, and expensive. This project attempts to bridge the communication gap by designing a portable glove that captures the user’s ASL gestures and outputs the translated text on a smartphone. The glove is equipped with flex sensors, contact sensors, and a gyroscope to measure the flexion of the fingers, the contact between fingers, and the rotation of the hand. The glove’s Arduino UNO microcontroller analyzes the sensor readings to identify the gesture from a library of learned gestures. The Bluetooth module transmits the gesture to a smartphone. Using this device, one day speakers of ASL may be able to communicate with others in an affordable and convenient way.

Keywords: sign language, morse code, convolutional neural network, American sign language, gesture recognition

Procedia PDF Downloads 63

Authors: Reza Mohammadzadeh

Abstract:

The major challenges in geotechnical engineering in underground spaces arise from uncertainties and different probabilities. The collection, collation, and collaboration of existing data to incorporate them in analysis and design for given prospect evaluation would be a reliable, practical problem solving method under uncertainty. Machine learning (ML) is a subfield of artificial intelligence in statistical science which applies different techniques (e.g., Regression, neural networks, support vector machines, decision trees, random forests, genetic programming, etc.) on data to automatically learn and improve from them without being explicitly programmed and make decisions and predictions. In this paper, a conceptual database schema of geotechnical risks in underground coal mining based on a cloud system architecture has been designed. A new approach of risk assessment using a three-dimensional risk matrix supported by the level of knowledge (LoK) has been proposed in this model. Subsequently, the model workflow methodology stages have been described. In order to train data and LoK models deployment, an ML platform has been implemented. IBM Watson Studio, as a leading data science tool and data-driven cloud integration ML platform, is employed in this study. As a Use case, a data set of geotechnical hazards and risk assessment in underground coal mining were prepared to demonstrate the performance of the model, and accordingly, the results have been outlined.

Keywords: data model, geotechnical risks, machine learning, underground coal mining

Procedia PDF Downloads 274

Authors: Seyoung Kim, Jeongmin Kim, Kwang Ryel Ryu

Abstract:

A database that records average traffic speeds measured at five-minute intervals for all the links in the traffic network of a metropolitan city. While learning from this data the models that can predict future traffic speed would be beneficial for the applications such as the car navigation system, building predictive models for every link becomes a nontrivial job if the number of links in a given network is huge. An advantage of adopting k-nearest neighbor (k-NN) as predictive models is that it does not require any explicit model building. Instead, k-NN takes a long time to make a prediction because it needs to search for the k-nearest neighbors in the database at prediction time. In this paper, we investigate how much we can speed up k-NN in making traffic speed predictions by reducing the amount of data to be searched for without a significant sacrifice of prediction accuracy. The rationale behind this is that we had a better look at only the recent data because the traffic patterns not only repeat daily or weekly but also change over time. In our experiments, we build several different k-NN models employing different sets of features which are the current and past traffic speeds of the target link and the neighbor links in its up/down-stream. The performances of these models are compared by measuring the average prediction accuracy and the average time taken to make a prediction using various amounts of data.

Keywords: big data, k-NN, machine learning, traffic speed prediction

Procedia PDF Downloads 363

Authors: Byung-Ik Kim, Hong-Koo Kang, Tae-Jin Lee, Hae-Ryong Park

Abstract:

Cyber terrors against specific enterprises or countries have been increasing recently. Such attacks against specific targets are called advanced persistent threat (APT), and they are giving rise to serious social problems. The malicious behaviors of APT attacks mostly affect websites and penetrate enterprise networks to perform malevolent acts. Although many enterprises invest heavily in security to defend against such APT threats, they recognize the APT attacks only after the latter are already in action. This paper discusses the characteristics of APT attacks at each step as well as the strengths and weaknesses of existing malicious code detection technologies to check their suitability for detecting APT attacks. It then proposes a network-based malicious behavior detection algorithm to protect the enterprise or national networks.

Keywords: Advanced Persistent Threat (APT), malware, network security, network packet, exploit kits

Procedia PDF Downloads 366

Authors: Simiao Ren, En Wei

Abstract:

Drowning is a significant safety issue worldwide, and a robust computer vision-based alert system can easily prevent such tragedies in swimming pools. However, due to domain shift caused by the visual gap (potentially due to lighting, indoor scene change, pool floor color etc.) between the training swimming pool and the test swimming pool, the robustness of such algorithms has been questionable. The annotation cost for labeling each new swimming pool is too expensive for mass adoption of such a technique. To address this issue, we propose a domain-aware data augmentation pipeline based on Gaussian Poisson Generative Adversarial Network (GP-GAN). Combined with YOLOv8, we demonstrate that such a domain adaptation technique can significantly improve the model performance (from 0.24 mAP to 0.82 mAP) on new test scenes. As the augmentation method only require background imagery from the new domain (no annotation needed), we believe this is a promising, practical route for preventing swimming pool drowning.

Keywords: computer vision, deep learning, YOLOv8, detection, swimming pool, drowning, domain adaptation, generative adversarial network, GAN, GP-GAN

Procedia PDF Downloads 101

Authors: C. Mayr, J. Periya, A. Kariminezhad

Abstract:

In this paper, we developed a learning framework for the classification of vulnerable road users (VRU) by their range-Doppler signatures. The frequency-modulated continuous-wave (FMCW) radar raw data is first pre-processed to obtain robust object range-Doppler maps per coherent time interval. The complex-valued range-Doppler maps captured from our outdoor measurements are further fed into a convolutional neural network (CNN) to learn the classification. This CNN has gone through a hyperparameter optimization process for improved learning. By learning VRU range-Doppler signatures, the three classes 'pedestrian', 'dog', and 'noise' are classified with an average accuracy of almost 95%. Interestingly, this classification accuracy holds for a combined longitudinal and lateral object trajectories.

Keywords: machine learning, radar, signal processing, autonomous driving

Procedia PDF Downloads 246

Authors: Wann-Ming Wey

Abstract:

In today's popularity and progress of information technology, the big data set and its analysis are no longer a major conundrum. Now, we could not only use the relevant big data to analysis and emulate the possible status of urban development in the near future, but also provide more comprehensive and reasonable policy implementation basis for government units or decision-makers via the analysis and emulation results as mentioned above. In this research, we set Taipei City as the research scope, and use the relevant big data variables (e.g., population, facility utilization and related social policy ratings) and Analytic Network Process (ANP) approach to implement in-depth research and discussion for the possible reduction of land use in primary and secondary schools of Taipei City. In addition to enhance the prosperous urban activities for the urban public facility utilization, the final results of this research could help improve the efficiency of urban land use in the future. Furthermore, the assessment model and research framework established in this research also provide a good reference for schools or other public facilities land use and adaptive reuse strategies in the future.

Keywords: adaptive reuse, analytic network process, big data, land use strategy

Procedia PDF Downloads 203

Authors: Suchitra Sivakumar, Hajime Shingyouchi, Toshinori Okajima, Kyohei Yamaguchi, Jin Kusaka

Abstract:

The progressing need for powertrain electrification calls for more accurate and reliable simulation models. A battery pack serves as the most vital component for energy storage in an electrified powertrain. Hybrid electric vehicles (HEV) do not behave the same way as they age, and there are several environmental factors that account for the degradation of the battery on a system level. Therefore, in this work, a battery model was proposed to study the state of charge (SOC) variation and the internal dynamic changes that contribute to aging and performance degradation in HEV batteries. An equivalent circuit battery model (ECM) is built using MATLAB Simulink to investigate the output characteristics of the lithium-ion battery. The ECM comprises of circuit elements like a voltage source, a series resistor and a parallel RC network connected in series. A parameter estimation study is conducted on the ECM to study the dependencies of the circuit elements with the state of charge (SOC) and the terminal voltage of the battery. The battery model is extended to simulate the temperature dependence of the individual battery cell and the battery pack with the environment. The temperature dependence model accounts for the heat loss due to internal resistance build up in the battery pack during charging, discharging, and due to atmospheric temperature. The model was validated for a lithium-ion battery pack with an independent drive cycle showing a voltage accuracy of 4% and SOC accuracy of about 2%.

Keywords: battery model, hybrid electric vehicle, lithium-ion battery, thermal model

Procedia PDF Downloads 298

Authors: Gheimati Salar

Abstract:

The objective of conducting this research was to study the influence of plyometric exercises on the biomechanical selected factor of elite teen swimmers and compare the influence of plyometric exercises on the speed of front crawl and back crawl in empirical and control groups of teens. In order to conduct these study 30 swimmers with minimum of 3 years' experience who were 11 or 12 were randomly chosen and divided into 2 groups of 15. The first group was empirical and the second was control group. The speed of the swimmer was analyzed after 25 meters of swimming and their speed were recorded in the last. The researcher was standing stable at the beginning and then started the chronometer and stopped it at the end of the swimming. He repeated the record taking two times and then the average was taken. Before conducting plyometric exercises, a speed test was taken from both groups in both types of swimming. The duration of plyometric exercises was 8 weeks, every week 3 sessions and 24 sessions in total. The exercises in this study were focused on 3 parts of the body. Upper limb part, the lower part of the body and trunk area. Upper limb exercises consisted of four parts. The lower limb exercises consisted of 5 parts, and the trunk exercises consisted of four sections. A Medicine ball, cone and different weights were used in these exercises.

Keywords: plyometric, exercises, front crawl and back crawl, speed

Procedia PDF Downloads 124

Authors: Adel. M. Sharaf, Foad H. Gandoman

Abstract:

Nowadays, Photovoltaic-PV Farms/ Parks and large PV-Smart Grid Interface Schemes are emerging and commonly utilized in Renewable Energy distributed generation. However, PV-hybrid-Dc-Ac Schemes using interface power electronic converters usually has negative impact on power quality and stabilization of modern electrical network under load excursions and network fault conditions in smart grid. Consequently, robust FACTS based interface schemes are required to ensure efficient energy utilization and stabilization of bus voltages as well as limiting switching/fault onrush current condition. FACTS devices are also used in smart grid-Battery Interface and Storage Schemes with PV-Battery Storage hybrid systems as an elegant alternative to renewable energy utilization with backup battery storage for electric utility energy and demand side management to provide needed energy and power capacity under heavy load conditions. The paper presents a robust interface PV-Li-Ion Battery Storage Interface Scheme for Distribution/Utilization Low Voltage Interface using FACTS stabilization enhancement and dynamic maximum PV power tracking controllers. Digital simulation and validation of the proposed scheme is done using MATLAB/Simulink software environment for Low Voltage- Distribution/Utilization system feeding a hybrid Linear-Motorized inrush and nonlinear type loads from a DC-AC Interface VSC-6-pulse Inverter Fed from the PV Park/Farm with a back-up Li-Ion Storage Battery.

Keywords: AC FACTS, smart grid, stabilization, PV-battery storage, Switched Filter-Compensation (SFC)

Procedia PDF Downloads 412

Authors: Jonathan Martino, Kristof Harri

Abstract:

In some cases, the increase in computing resources makes simulation methods more affordable. The increase in processing speed also allows real time analysis or even more rapid tests analysis offering a real tool for test prediction and design process optimization. Vibration tests are no exception to this trend. The so called ‘Virtual Vibration Testing’ offers solution among others to study the influence of specific loads, to better anticipate the boundary conditions between the exciter and the structure under test, to study the influence of small changes in the structure under test, etc. This article will first present a virtual vibration test modeling with a main focus on the shaker model and will afterwards present the experimental parameters determination. The classical way of modeling a shaker is to consider the shaker as a simple mechanical structure augmented by an electrical circuit that makes the shaker move. The shaker is modeled as a two or three degrees of freedom lumped parameters model while the electrical circuit takes the coil impedance and the dynamic back-electromagnetic force into account. The establishment of the equations of this model, describing the dynamics of the shaker, is presented in this article and is strongly related to the internal physical quantities of the shaker. Those quantities will be reduced into global parameters which will be estimated through experiments. Different experiments will be carried out in order to design an easy and practical method for the identification of the shaker parameters leading to a fully functional shaker model. An experimental modal analysis will also be carried out to extract the modal parameters of the shaker and to combine them with the electrical measurements. Finally, this article will conclude with an experimental validation of the model.

Keywords: lumped parameters model, shaker modeling, shaker parameters, state-space, virtual vibration

Procedia PDF Downloads 270

Authors: Juhi Kaushik, Abhishek Agarwal, Manoj Sarda, Vatsal Sanjay, Arup Kumar Das

Abstract:

Fire accidents in trains bring huge disaster to human life and property. Evacuation becomes a major challenge in such incidents owing to confined spaces, large passenger density and trains moving at high speeds. The pantry car in Indian Railways trains carry inflammable materials like cooking fuel and LPG and electrical fittings. The pantry car is therefore highly susceptible to fire accidents. Numerical simulations have been done in a pantry car of Indian locomotive train using computational fluid dynamics based software. Different scenarios of a fire outbreak have been explored by varying Heat Release Rate per Unit Area (HRRPUA) of the fire source, introduction of exhaust in the cooking area, and taking a case of an air conditioned pantry car. Temporal statures of flame and soot have been obtained for each scenario and differences have been studied and reported. Inputs from this study can be used to assess casualties in fire accidents in locomotive trains and development of smoke control/detection systems in Indian trains.

Keywords: fire propagation, flame contour, pantry fire, soot flow

Procedia PDF Downloads 339