Search results for: physics-informed neural networks

Authors: David Percy

Abstract:

Landslides are a geologic phenomenon that affects a large number of inhabited places and are constantly being monitored and studied for the prediction of future occurrences. Reconstructability analysis (RA) is a methodology for extracting informative models from large volumes of data that work exclusively with discrete data. While RA has been used in medical applications and social science extensively, we are introducing it to the spatial sciences through applications like landslide prediction. Since RA works exclusively with discrete data, such as soil classification or bedrock type, working with continuous data, such as porosity, requires that these data are binned for inclusion in the model. RA constructs models of the data which pick out the most informative elements, independent variables (IVs), from each layer that predict the dependent variable (DV), landslide occurrence. Each layer included in the model retains its classification data as a primary encoding of the data. Unlike other machine learning algorithms that force the data into one-hot encoding type of schemes, RA works directly with the data as it is encoded, with the exception of continuous data, which must be binned. The usual physical and derived layers are included in the model, and testing our results against other published methodologies, such as neural networks, yields accuracy that is similar but with the advantage of a completely transparent model. The results of an RA session with a data set are a report on every combination of variables and their probability of landslide events occurring. In this way, every combination of informative state combinations can be examined.

Keywords: reconstructability analysis, machine learning, landslides, raster analysis

Procedia PDF Downloads 66

Authors: Dongkwon Han, Sangho Kim, Sunil Kwon

Abstract:

Unconventional gas resources have dramatically changed the future energy landscape. Unlike conventional gas resources, the key challenges in unconventional gas have been the requirement that applies to advanced approaches for production forecasting due to uncertainty and complexity of fluid flow. In this study, artificial neural network (ANN) model which integrates machine learning and data-driven approach was developed to predict productivity in shale gas. The database of 129 wells of Eagle Ford shale basin used for testing and training of the ANN model. The Input data related to hydraulic fracturing, well completion and productivity of shale gas were selected and the output data is a cumulative production. The performance of the ANN using all data sets, clustering and variables importance (VI) models were compared in the mean absolute percentage error (MAPE). ANN model using all data sets, clustering, and VI were obtained as 44.22%, 10.08% (cluster 1), 5.26% (cluster 2), 6.35%(cluster 3), and 32.23% (ANN VI), 23.19% (SVM VI), respectively. The results showed that the pre-trained ANN model provides more accurate results than the ANN model using all data sets.

Keywords: unconventional gas, artificial neural network, machine learning, clustering, variables importance

Procedia PDF Downloads 196

Authors: Farshid Hajati, Federico Girosi, Shima Ghassempour

Abstract:

We present a Deep Belief Network (DBN) method for clustering health trajectories. Deep Belief Network (DBN) is a deep architecture that consists of a stack of Restricted Boltzmann Machines (RBM). In a deep architecture, each layer learns more complex features than the past layers. The proposed method depends on DBN in clustering without using back propagation learning algorithm. The proposed DBN has a better a performance compared to the deep neural network due the initialization of the connecting weights. We use Contrastive Divergence (CD) method for training the RBMs which increases the performance of the network. The performance of the proposed method is evaluated extensively on the Health and Retirement Study (HRS) database. The University of Michigan Health and Retirement Study (HRS) is a nationally representative longitudinal study that has surveyed more than 27,000 elderly and near-elderly Americans since its inception in 1992. Participants are interviewed every two years and they collect data on physical and mental health, insurance coverage, financial status, family support systems, labor market status, and retirement planning. The dataset is publicly available and we use the RAND HRS version L, which is easy to use and cleaned up version of the data. The size of sample data set is 268 and the length of the trajectories is equal to 10. The trajectories do not stop when the patient dies and represent 10 different interviews of live patients. Compared to the state-of-the-art benchmarks, the experimental results show the effectiveness and superiority of the proposed method in clustering health trajectories.

Keywords: health trajectory, clustering, deep learning, DBN

Procedia PDF Downloads 369

Authors: J. K. Adedeji, M. O. Oyekanmi

Abstract:

This paper has critically examined the use of Machine Learning procedures in curbing unauthorized access into valuable areas of an organization. The use of passwords, pin codes, user’s identification in recent times has been partially successful in curbing crimes involving identities, hence the need for the design of a system which incorporates biometric characteristics such as DNA and pattern recognition of variations in facial expressions. The facial model used is the OpenCV library which is based on the use of certain physiological features, the Raspberry Pi 3 module is used to compile the OpenCV library, which extracts and stores the detected faces into the datasets directory through the use of camera. The model is trained with 50 epoch run in the database and recognized by the Local Binary Pattern Histogram (LBPH) recognizer contained in the OpenCV. The training algorithm used by the neural network is back propagation coded using python algorithmic language with 200 epoch runs to identify specific resemblance in the exclusive OR (XOR) output neurons. The research however confirmed that physiological parameters are better effective measures to curb crimes relating to identities.

Keywords: biometric characters, facial recognition, neural network, OpenCV

Procedia PDF Downloads 256

Authors: Bircan Demiral

Abstract:

Cognitive radio (CR) is the promising technology that addresses the spectrum scarcity problem for future wireless communications. Orthogonal Frequency Division Multiplexing (OFDM) technology provides more power band ratios for cognitive radio networks (CRNs). While CR is a solution to the spectrum scarcity, it also brings up the capacity problem. In this paper, a novel power allocation algorithm that aims at maximizing the sum capacity in the OFDM based cognitive radio networks is proposed. Proposed allocation algorithm is based on the previously developed water-filling algorithm. To reduce the computational complexity calculating in water filling algorithm, proposed algorithm allocates the total power according to each subcarrier. The power allocated to the subcarriers increases sum capacity. To see this increase, Matlab program was used, and the proposed power allocation was compared with average power allocation, water filling and general power allocation algorithms. The water filling algorithm performed worse than the proposed algorithm while it performed better than the other two algorithms. The proposed algorithm is better than other algorithms in terms of capacity increase. In addition the effect of the change in the number of subcarriers on capacity was discussed. Simulation results show that the increase in the number of subcarrier increases the capacity.

Keywords: cognitive radio network, OFDM, power allocation, water filling

Procedia PDF Downloads 137

Authors: Yeliz Karaca, Carlo Cattani

Abstract:

This study focuses on the local Holder exponent as a measure of the function regularity for time series related to finance data. In this study, the attributes of the finance dataset belonging to 13 countries (India, China, Japan, Sweden, France, Germany, Italy, Australia, Mexico, United Kingdom, Argentina, Brazil, USA) located in 5 different continents (Asia, Europe, Australia, North America and South America) have been examined.These countries are the ones mostly affected by the attributes with regard to financial development, covering a period from 2012 to 2017. Our study is concerned with the most important attributes that have impact on the development of finance for the countries identified. Our method is comprised of the following stages: (a) among the multi fractal methods and Brownian motion Holder regularity functions (polynomial, exponential), significant and self-similar attributes have been identified (b) The significant and self-similar attributes have been applied to the Artificial Neuronal Network (ANN) algorithms (Feed Forward Back Propagation (FFBP) and Cascade Forward Back Propagation (CFBP)) (c) the outcomes of classification accuracy have been compared concerning the attributes that have impact on the attributes which affect the countries’ financial development. This study has enabled to reveal, through the application of ANN algorithms, how the most significant attributes are identified within the relevant dataset via the Holder functions (polynomial and exponential function).

Keywords: artificial neural networks, finance data, Holder regularity, multifractals

Procedia PDF Downloads 246

Authors: Parisa Mansour

Abstract:

Understanding how vision works in both health and disease involves understanding the anatomy and physiology of the eye as well as the neural pathways involved in visual perception. The development of imaging techniques for the visual system is essential for understanding the neural foundation of visual function or impairment. MRI provides a way to examine neural circuit structure and function without invasive procedures, allowing for the detection of brain tissue abnormalities in real time. One of the advanced MRI methods is manganese-enhanced MRI (MEMRI), which utilizes active manganese contrast agents to enhance brain tissue signals in T1-weighted imaging, showcasing connectivity and activity levels. The way manganese ions build up in the eye, and visual pathways can be due to their spread throughout the body or by moving locally along axons in a forward direction and entering neurons through calcium channels that are voltage-gated. The paramagnetic manganese contrast is utilized in MRI for various applications in the visual system, such as imaging neurodevelopment and evaluating neurodegeneration, neuroplasticity, neuroprotection, and neuroregeneration. In this assessment, we outline four key areas of scientific research where MEMRI can play a crucial role - understanding brain structure, mapping nerve pathways, monitoring nerve cell function, and distinguishing between different types of glial cell activity. We discuss various studies that have utilized MEMRI to investigate the visual system, including delivery methods, spatiotemporal features, and biophysical analysis. Based on this literature, we have pinpointed key issues in the field related to toxicity, as well as sensitivity and specificity of manganese enhancement. We will also examine the drawbacks and other options to MEMRI that could offer new possibilities for future exploration.

Keywords: glial activity, manganese-enhanced magnetic resonance imaging, neuroarchitecture, neuronal activity, neuronal tract tracing, visual pathway, eye

Procedia PDF Downloads 40

Authors: Norbert Magyar, Jozsef Kovacs, Peter Tanos, Balazs Trasy, Tamas Garamhegyi, Istvan Gabor Hatvani

Abstract:

Water is one of the most important common resources, and as a result of urbanization, agriculture, and industry it is becoming more and more exposed to potential pollutants. The prevention of the deterioration of water quality is a crucial role for environmental scientist. To achieve this aim, the operation of monitoring networks is necessary. In general, these networks have to meet many important requirements, such as representativeness and cost efficiency. However, existing monitoring networks often include sampling sites which are unnecessary. With the elimination of these sites the monitoring network can be optimized, and it can operate more economically. The aim of this study is to illustrate the applicability of the CCDA (Combined Cluster and Discriminant Analysis) to the field of water quality monitoring and optimize the monitoring networks of a river (the Danube), a wetland-lake system (Kis-Balaton & Lake Balaton), and two surface-subsurface water systems on the watershed of Lake Neusiedl/Lake Fertő and on the Szigetköz area over a period of approximately two decades. CCDA combines two multivariate data analysis methods: hierarchical cluster analysis and linear discriminant analysis. Its goal is to determine homogeneous groups of observations, in our case sampling sites, by comparing the goodness of preconceived classifications obtained from hierarchical cluster analysis with random classifications. The main idea behind CCDA is that if the ratio of correctly classified cases for a grouping is higher than at least 95% of the ratios for the random classifications, then at the level of significance (α=0.05) the given sampling sites don’t form a homogeneous group. Due to the fact that the sampling on the Lake Neusiedl/Lake Fertő was conducted at the same time at all sampling sites, it was possible to visualize the differences between the sampling sites belonging to the same or different groups on scatterplots. Based on the results, the monitoring network of the Danube yields redundant information over certain sections, so that of 12 sampling sites, 3 could be eliminated without loss of information. In the case of the wetland (Kis-Balaton) one pair of sampling sites out of 12, and in the case of Lake Balaton, 5 out of 10 could be discarded. For the groundwater system of the catchment area of Lake Neusiedl/Lake Fertő all 50 monitoring wells are necessary, there is no redundant information in the system. The number of the sampling sites on the Lake Neusiedl/Lake Fertő can decrease to approximately the half of the original number of the sites. Furthermore, neighbouring sampling sites were compared pairwise using CCDA and the results were plotted on diagrams or isoline maps showing the location of the greatest differences. These results can help researchers decide where to place new sampling sites. The application of CCDA proved to be a useful tool in the optimization of the monitoring networks regarding different types of water bodies. Based on the results obtained, the monitoring networks can be operated more economically.

Keywords: combined cluster and discriminant analysis, cost efficiency, monitoring network optimization, water quality

Procedia PDF Downloads 348

Authors: S. Abera, T. Gidey, W. Terefe

Abstract:

Introduction: Testing for HIV is the key entry point to HIV prevention, treatment, and care and support services. Hence, predictive data mining techniques can greatly benefit to analyze and discover new patterns from huge datasets like that of EDHS 2011 data. Objectives: The objective of this study is to build a predictive modeling for HIV testing and identify determinants of HIV testing for adults with age above fourteen years using data mining techniques. Methods: Cross-Industry Standard Process for Data Mining (CRISP-DM) was used to predict the model for HIV testing and explore association rules between HIV testing and the selected attributes among adult Ethiopians. Decision tree, Naïve-Bayes, logistic regression and artificial neural networks of data mining techniques were used to build the predictive models. Results: The target dataset contained 30,625 study participants; of which 16, 515 (53.9%) were women. Nearly two-fifth; 17,719 (58%), have never been tested for HIV while the rest 12,906 (42%) had been tested. Ethiopians with higher wealth index, higher educational level, belonging 20 to 29 years old, having no stigmatizing attitude towards HIV positive person, urban residents, having HIV related knowledge, information about family planning on mass media and knowing a place where to get testing for HIV showed an increased patterns with respect to HIV testing. Conclusion and Recommendation: Public health interventions should consider the identified determinants to promote people to get testing for HIV.

Keywords: data mining, HIV, testing, ethiopia

Procedia PDF Downloads 496

Authors: Laheeb M. Ibrahim, Ibrahim A. Salih

Abstract:

Biometrics is a technology that has been widely used in many official and commercial identification applications. The increased concerns in security during recent years (especially during the last decades) have essentially resulted in more attention being given to biometric-based verification techniques. Here, a novel fusion approach of palmprint, dental traits has been suggested. These traits which are authentication techniques have been employed in a range of biometric applications that can identify any postmortem PM person and antemortem AM. Besides improving the accuracy, the fusion of biometrics has several advantages such as increasing, deterring spoofing activities and reducing enrolment failure. In this paper, a first unimodel biometric system has been made by using (palmprint and dental) traits, for each one classification applying an artificial neural network and a hybrid technique that combines swarm intelligence and neural network together, then attempt has been made to combine palmprint and dental biometrics. Principally, the fusion of palmprint and dental biometrics and their potential application has been explored as biometric identifiers. To address this issue, investigations have been carried out about the relative performance of several statistical data fusion techniques for integrating the information in both unimodal and multimodal biometrics. Also the results of the multimodal approach have been compared with each one of these two traits authentication approaches. This paper studies the features and decision fusion levels in multimodal biometrics. To determine the accuracy of GAR to parallel system decision-fusion including (AND, OR, Majority fating) has been used. The backpropagation method has been used for classification and has come out with result (92%, 99%, 97%) respectively for GAR, while the GAR) for this algorithm using hybrid technique for classification (95%, 99%, 98%) respectively. To determine the accuracy of the multibiometric system for feature level fusion has been used, while the same preceding methods have been used for classification. The results have been (98%, 99%) respectively while to determine the GAR of feature level different methods have been used and have come out with (98%).

Keywords: back propagation neural network BP ANN, multibiometric system, parallel system decision-fusion, practical swarm intelligent PSO

Procedia PDF Downloads 533

Authors: Ayalew Tadese Kibret, Belayneh Sisay Alemu, Amare Kassaw Yimer

Abstract:

Intelligent reflecting surfaces (IRSs) are considered to be a key enabling technology for sixth-generation (6G) wireless networks. IRSs are electromagnetic (EM) surfaces that are fabricated and have integrated electronics, electronically controlled processes, and particularly wireless communication features. IRSs operate without the need for complex signal processing and the encoding and decoding steps that improve the signal quality at the receiver. Improving vital performance parameters such as energy efficiency (EE) and spectral efficiency (SE) have frequently been the primary goals of research in order to meet the increasing requirements for advanced services in the future 6G communications. In this research, we conduct a comparative analysis on single and multi-IRS wireless communication networks using energy and spectrum efficiency. The energy efficiency versus user distance, energy efficiency versus signal to noise ratio, and spectral efficiency versus user distance are the basis for our result with 1, 2, 4, and 6 IRSs. According to the results of our simulation, in terms of energy and spectral efficiency, six IRS perform better than four, two, and single IRS. Overall, our results suggest that multi-IRS-assisted wireless communication systems outperform single IRS systems in terms of communication performance.

Keywords: sixth-generation (6G), wireless networks, intelligent reflecting surfaces, energy efficiency, spectral efficiency

Procedia PDF Downloads 25

Authors: Mahmoud Samkan

Abstract:

Dispersed Generation (DG) is a promising solution to many power system problems such as voltage regulation and power loss. This paper proposes a heuristic two-step method to optimize the location and size of DG for reducing active power losses and, therefore, improve the voltage profile in radial distribution networks. In addition to a DG placed at the system load gravity center, this method consists in assigning a DG to each lateral of the network. After having determined the central DG placement, the location and size of each lateral DG are predetermined in the first step. The results are then refined in the second step. This method is tested for 33-bus system for 100% DG penetration. The results obtained are compared with those of other methods found in the literature.

Keywords: optimal location, optimal size, dispersed generation (DG), radial distribution networks, reducing losses

Procedia PDF Downloads 443

Authors: Wei-Han Wang, Hsiang-Yu Yu, Mau-Sun Hua

Abstract:

Theory of Mind (ToM) refers to the ability to infer another’s mental state. With appropriate ToM, one can behave well in social interactions. A growing body of evidence has demonstrated that patients with temporal lobe epilepsy (TLE) may have damaged ToM due to impact on regions of the underlying neural network of ToM. However, the question of whether there is cerebral laterality for ToM functions remains open. This study aimed to examine whether there is cerebral lateralization for ToM abilities in TLE patients. Sixty-seven adult TLE patients and 30 matched healthy controls (HC) were recruited. Patients were classified into right (RTLE), left (LTLE), and bilateral (BTLE) TLE groups on the basis of a consensus panel review of their seizure semiology, EEG findings, and brain imaging results. All participants completed an intellectual test and four tasks measuring basic and advanced ToM. The results showed that, on all ToM tasks; (1)each patient group performed worse than HC; (2)there were no significant differences between LTLE and RTLE groups; (3)the BTLE group performed the worst. It appears that the neural network responsible for ToM is distributed evenly between the cerebral hemispheres.

Keywords: cerebral lateralization, social cognition, temporal lobe epilepsy, theory of mind

Procedia PDF Downloads 420

Authors: Halefom Kidane

Abstract:

This study aims to assess the wind resource potential and characterize the urban area wind patterns in Hawassa City, Ethiopia. The estimation and characterization of wind resources are crucial for sustainable urban planning, renewable energy development, and climate change mitigation strategies. A secondary data collection method was used to carry out the study. The collected data at 2 meters was analyzed statistically and extrapolated to the standard heights of 10-meter and 30-meter heights using the power law equation. The standard deviation method was used to calculate the value of scale and shape factors. From the analysis presented, the maximum and minimum mean daily wind speed at 2 meters in 2016 was 1.33 m/s and 0.05 m/s in 2017, 1.67 m/s and 0.14 m/s in 2018, 1.61m and 0.07 m/s, respectively. The maximum monthly average wind speed of Hawassa City in 2016 at 2 meters was noticed in the month of December, which is around 0.78 m/s, while in 2017, the maximum wind speed was recorded in the month of January with a wind speed magnitude of 0.80 m/s and in 2018 June was maximum speed which is 0.76 m/s. On the other hand, October was the month with the minimum mean wind speed in all years, with a value of 0.47 m/s in 2016,0.47 in 2017 and 0.34 in 2018. The annual mean wind speed was 0.61 m/s in 2016,0.64, m/s in 2017 and 0.57 m/s in 2018 at a height of 2 meters. From extrapolation, the annual mean wind speeds for the years 2016,2017 and 2018 at 10 heights were 1.17 m/s,1.22 m/s, and 1.11 m/s, and at the height of 30 meters, were 3.34m/s,3.78 m/s, and 3.01 m/s respectively/Thus, the site consists mainly primarily classes-I of wind speed even at the extrapolated heights.

Keywords: artificial neural networks, forecasting, min-max normalization, wind speed

Procedia PDF Downloads 75

Authors: Akash Amalan, Rui Wang, Yanqi Qiao, Emmanouil Panaousis, Kaitai Liang

Abstract:

Federated learning is an emerging concept in the domain of distributed machine learning. This concept has enabled General Adversarial Networks (GANs) to benefit from the rich distributed training data while preserving privacy. However, in a non-IID setting, current federated GAN architectures are unstable, struggling to learn the distinct features, and vulnerable to mode collapse. In this paper, we propose an architecture MULTI-FLGAN to solve the problem of low-quality images, mode collapse, and instability for non-IID datasets. Our results show that MULTI-FLGAN is four times as stable and performant (i.e., high inception score) on average over 20 clients compared to baseline FLGAN.

Keywords: federated learning, generative adversarial network, inference attack, non-IID data distribution

Procedia PDF Downloads 158

Authors: Ninlawat Phuangchoke, Waraporn Viyanon, Setta Sasananan

Abstract:

The most important process of the water treatment plant process is the coagulation using alum and poly aluminum chloride (PACL), and the value of usage per day is a hundred thousand baht. Therefore, determining the dosage of alum and PACL are the most important factors to be prescribed. Water production is economical and valuable. This research applies an artificial neural network (ANN), which uses the Levenberg–Marquardt algorithm to create a mathematical model (Soft Jar Test) for prediction chemical dose used to coagulation such as alum and PACL, which input data consists of turbidity, pH, alkalinity, conductivity, and, oxygen consumption (OC) of Bangkhen water treatment plant (BKWTP) Metropolitan Waterworks Authority. The data collected from 1 January 2019 to 31 December 2019 cover changing seasons of Thailand. The input data of ANN is divided into three groups training set, test set, and validation set, which the best model performance with a coefficient of determination and mean absolute error of alum are 0.73, 3.18, and PACL is 0.59, 3.21 respectively.

Keywords: soft jar test, jar test, water treatment plant process, artificial neural network

Procedia PDF Downloads 166

Authors: Nadia Masood Khan, Muhammad Salman Khan, Gul Muhammad Khan

Abstract:

Cardiologists perform cardiac auscultation to detect abnormalities in heart sounds. Since accurate auscultation is a crucial first step in screening patients with heart diseases, there is a need to develop computer-aided detection/diagnosis (CAD) systems to assist cardiologists in interpreting heart sounds and provide second opinions. In this paper different algorithms are implemented for automated heart sound classification using unsegmented phonocardiogram (PCG) signals. Support vector machine (SVM), artificial neural network (ANN) and cartesian genetic programming evolved artificial neural network (CGPANN) without the application of any segmentation algorithm has been explored in this study. The signals are first pre-processed to remove any unwanted frequencies. Both time and frequency domain features are then extracted for training the different models. The different algorithms are tested in multiple scenarios and their strengths and weaknesses are discussed. Results indicate that SVM outperforms the rest with an accuracy of 73.64%.

Keywords: pattern recognition, machine learning, computer aided diagnosis, heart sound classification, and feature extraction

Procedia PDF Downloads 263

Authors: Sajjad Baghernezhad, Saeideh Baghernezhad

Abstract:

Vast changes and developments achieved in information technology field in recent decades have made the review of different issues such as organizational structures unavoidable. Applying informative technologies such as internet and also vast use of computer and related networks have led to new organizational formations with a nature completely different from the traditional, great and bureaucratic ones; some common specifications of such organizations are transfer of the affairs out of the organization, benefiting from informative and communicative networks and centered-science workers. Such communicative necessities have led to network sciences development including grid computing. First, the grid computing was only to relate some sites for short – time and use their sources simultaneously, but now it has gone beyond such idea. In this article, the grid computing technology was examined, and at the same time, virtual organization concept was discussed.

Keywords: grid computing, virtual organizations, software engineering, organization

Procedia PDF Downloads 332

Authors: Sekkal Nawel

Abstract:

The Web of Things (WoT), integration of smart technologies from the Internet or network to Web architecture or application, is becoming more complex, larger, and dynamic. The WoT is associated with various elements such as sensors, devices, networks, protocols, data, functionalities, and architectures to perform services for stakeholders. These services operate in the context of the interaction of stakeholders and the WoT elements. Such context is becoming a key information source from which data are of various nature and uncertain, thus leading to complex situations. In this paper, we take interest in the development of intelligent Web services. The key ingredients of this “intelligent” notion are the context diversity, the necessity of a semantic representation to manage complex situations and the capacity to reason with uncertain data. In this perspective, we introduce a multi-layered architecture based on a generic intelligent Web service model dealing with various contexts, which proactively predict future situations and reactively respond to real-time situations in order to support decision-making. For semantic context data representation, we use PR-OWL, which is a probabilistic ontology based on Multi-Entity Bayesian Networks (MEBN). PR-OWL is flexible enough to represent complex, dynamic, and uncertain contexts, the key requirements of the development for the intelligent Web services. A case study was carried out using the proposed architecture for intelligent plant watering to show the role of proactive and reactive contextual reasoning in terms of WoT.

Keywords: smart web service, the web of things, context reasoning, proactive, reactive, multi-entity bayesian networks, PR-OWL

Procedia PDF Downloads 71

Authors: Huy Hieu Pham, Houssam Salmane, Louahdi Khoudour, Alain Crouzil, Pablo Zegers, Sergio Velastin

Abstract:

We present a deep learning-based multitask framework for joint 3D human pose estimation and action recognition from color video sequences. Our approach proceeds along two stages. In the first, we run a real-time 2D pose detector to determine the precise pixel location of important key points of the body. A two-stream neural network is then designed and trained to map detected 2D keypoints into 3D poses. In the second, we deploy the Efficient Neural Architecture Search (ENAS) algorithm to find an optimal network architecture that is used for modeling the Spatio-temporal evolution of the estimated 3D poses via an image-based intermediate representation and performing action recognition. Experiments on Human3.6M, Microsoft Research Redmond (MSR) Action3D, and Stony Brook University (SBU) Kinect Interaction datasets verify the effectiveness of the proposed method on the targeted tasks. Moreover, we show that our method requires a low computational budget for training and inference.

Keywords: human action recognition, pose estimation, D-CNN, deep learning

Procedia PDF Downloads 146

Authors: Simone Leonardi, Luca Ardito

Abstract:

Micro-influencers are effective elements in the marketing strategies of companies and institutions because of their capability to create an hyper-engaged audience around a specific topic of interest. In recent years, many scientific approaches and commercial tools have handled the task of detecting this type of social media users. These strategies adopt solutions ranging from rule based machine learning models to deep neural networks and graph analysis on text, images, and account information. This work compares the existing solutions and proposes an ensemble method to generalize them with different input data and social media platforms. The deployed solution combines deep learning models on unstructured data with statistical machine learning models on structured data. We retrieve both social media accounts information and multimedia posts on Twitter and Instagram. These data are mapped into feature vectors for an eXtreme Gradient Boosting (XGBoost) classifier. Sixty different topics have been analyzed to build a rule based gold standard dataset and to compare the performances of our approach against baseline classifiers. We prove the effectiveness of our work by comparing the accuracy, precision, recall, and f1 score of our model with different configurations and architectures. We obtained an accuracy of 0.91 with our best performing model.

Keywords: deep learning, gradient boosting, image processing, micro-influencers, NLP, social media

Procedia PDF Downloads 183

Authors: Cristina Chaminade, Mandar Kulkarni, Balaji Parthasarathy, Monica Plechero

Abstract:

The characteristics of a regional innovation system (RIS) and the specificity of the knowledge base of an industry may contribute to create peculiar paths for innovation and development of firms’ geographic extended innovation networks. However, the relative empirical evidence in emerging economies remains underexplored. The paper aims to fill the research gap by means of some recent qualitative research conducted in 2016 in Beijing (China) and Bangalore (India). It analyzes cases studies of firms in the new media industry, a sector that merges different IT competences with competences from other knowledge domains and that is emerging in those RIS. The results show that while in Beijing the new media sector results to be more in line with the existing institutional setting and governmental goals aimed at targeting specific social aspects and social problems of the population, in Bangalore it remains a more spontaneous firms-led process. In Beijing what matters for the development of innovation networks is the governmental setting and the national and regional strategies to promote science and technology in this sector, internet and mass innovation. The peculiarities of recent governmental policies aligned to the domestic goals may provide good possibilities for start-ups to develop innovation networks. However, due to the specificities of those policies targeting the Chinese market, networking outside the domestic market are not so promoted. Moreover, while some institutional peculiarities, such as a culture of collaboration in the region, may be favorable for local networking, regulations related to Internet censorship may limit the use of global networks particularly when based on virtual spaces. Mainly firms with already some foreign experiences and contact take advantage of global networks. In Bangalore, the role of government in pushing networking for the new media industry at the present stage is quite absent at all geographical levels. Indeed there is no particular strategic planning or prioritizing in the region toward the new media industry, albeit one industrial organization has emerged to represent the animation industry interests. This results in a lack of initiatives for sustaining the integration of complementary knowledge into the local portfolio of IT specialization. Firms actually involved in the new media industry face institutional constrains related to a poor level of local trust and cooperation, something that does not allow for full exploitation of local linkages. Moreover, knowledge-provider organizations in Bangalore remain still a solid base for the IT domain, but not for other domains. Initiatives to link to international networks seem therefore more the result of individual entrepreneurial actions aimed at acquiring complementary knowledge and competencies from different domains and exploiting potentiality in different markets. From those cases, it emerges that role of government, soft institutions and organizations in the two RIS differ substantially in the creation of barriers and opportunities for the development of innovation networks and their specific aim.

Keywords: regional innovation system, emerging economies, innovation network, institutions, organizations, Bangalore, Beijing

Procedia PDF Downloads 323

Authors: Elaine Lima Da Fonseca, Eliomar Pereira Da Silva Filho

Abstract:

The modeling of areas susceptible to soil loss by hydro erosive processes consists of a simplified instrument of reality with the purpose of predicting future behaviors from the observation and interaction of a set of geoenvironmental factors. The models of potential areas for soil loss will be obtained through binary logistic regression, artificial neural networks, and support vector machines. The choice of the municipality of Colorado do Oeste in the south of the western Amazon is due to soil degradation due to anthropogenic activities, such as agriculture, road construction, overgrazing, deforestation, and environmental and socioeconomic configurations. Initially, a soil erosion inventory map constructed through various field investigations will be designed, including the use of remotely piloted aircraft, orbital imagery, and the PLANAFLORO/RO database. 100 sampling units with the presence of erosion will be selected based on the assumptions indicated in the literature, and, to complement the dichotomous analysis, 100 units with no erosion will be randomly designated. The next step will be the selection of the predictive parameters that exert, jointly, directly, or indirectly, some influence on the mechanism of occurrence of soil erosion events. The chosen predictors are altitude, declivity, aspect or orientation of the slope, curvature of the slope, composite topographic index, flow power index, lineament density, normalized difference vegetation index, drainage density, lithology, soil type, erosivity, and ground surface temperature. After evaluating the relative contribution of each predictor variable, the erosion susceptibility model will be applied to the municipality of Colorado do Oeste - Rondônia through the SPSS Statistic 26 software. Evaluation of the model will occur through the determination of the values of the R² of Cox & Snell and the R² of Nagelkerke, Hosmer and Lemeshow Test, Log Likelihood Value, and Wald Test, in addition to analysis of the Confounding Matrix, ROC Curve and Accumulated Gain according to the model specification. The validation of the synthesis map resulting from both models of the potential risk of soil erosion will occur by means of Kappa indices, accuracy, and sensitivity, as well as by field verification of the classes of susceptibility to erosion using drone photogrammetry. Thus, it is expected to obtain the mapping of the following classes of susceptibility to erosion very low, low, moderate, very high, and high, which may constitute a screening tool to identify areas where more detailed investigations need to be carried out, applying more efficient social resources.

Keywords: modeling, susceptibility to erosion, artificial intelligence, Amazon

Procedia PDF Downloads 66

Authors: L. Rakesh, T. Y. Heblekar

Abstract:

This study entails the development and optimization of a mathematical model for a floating drum biogas reactor from first principles using thermal and empirical considerations. The model was derived on the basis of mass conservation, lumped mass heat transfer formulations and empirical biogas formation laws. The treatment leads to a system of coupled nonlinear ordinary differential equations whose solution mapped four-time independent controllable parameters to five output variables which adequately serve to describe the reactor performance. These equations were solved numerically using fourth order Runge-Kutta method for a range of input parameter values. Using the data so obtained an Artificial Neural Network with a single hidden layer was trained using Levenberg-Marquardt Damped Least Squares (DLS) algorithm. This network was then fine-tuned for optimal mapping by varying hidden layer size. This fast forward model was then employed as a health score generator in the Bacterial Foraging Optimization code. The optimal operating state of the simplified Biogas reactor was thus obtained.

Keywords: biogas, floating drum reactor, neural network model, optimization

Procedia PDF Downloads 143

Authors: Maik Kschischo, Dominik Kahl, Philipp Wendland, Andreas Weber

Abstract:

Understanding and modelling of real-world complex dynamic systems in biology, engineering and other fields is often made difficult by incomplete knowledge about the interactions between systems states and by unknown disturbances to the system. In fact, most real-world dynamic networks are open systems receiving unknown inputs from their environment. To understand a system and to estimate the state dynamics, these inputs need to be reconstructed from output measurements. Reconstructing the input of a dynamic system from its measured outputs is an ill-posed problem if only a limited number of states is directly measurable. A first requirement for solving this problem is the invertibility of the input-output map. In our work, we exploit the fact that invertibility of a dynamic system is a structural property, which depends only on the network topology. Therefore, it is possible to check for invertibility using a structural invertibility algorithm which counts the number of node disjoint paths linking inputs and outputs. The algorithm is efficient enough, even for large networks up to a million nodes. To understand structural features influencing the invertibility of a complex dynamic network, we analyze synthetic and real networks using the structural invertibility algorithm. We find that invertibility largely depends on the degree distribution and that dense random networks are easier to invert than sparse inhomogeneous networks. We show that real networks are often very difficult to invert unless the sensor nodes are carefully chosen. To overcome this problem, we present a sensor node placement algorithm to achieve invertibility with a minimum set of measured states. This greedy algorithm is very fast and also guaranteed to find an optimal sensor node-set if it exists. Our results provide a practical approach to experimental design for open, dynamic systems. Since invertibility is a necessary condition for unknown input observers and data assimilation filters to work, it can be used as a preprocessing step to check, whether these input reconstruction algorithms can be successful. If not, we can suggest additional measurements providing sufficient information for input reconstruction. Invertibility is also important for systems design and model building. Dynamic models are always incomplete, and synthetic systems act in an environment, where they receive inputs or even attack signals from their exterior. Being able to monitor these inputs is an important design requirement, which can be achieved by our algorithms for invertibility analysis and sensor node placement.

Keywords: data-driven dynamic systems, inversion of dynamic systems, observability, experimental design, sensor node placement

Procedia PDF Downloads 150

Authors: Bowen Ding, Yihao Kuang

Abstract:

In recent years, with the development of large-scale pre-trained lan-guage models, building vector representations of text through deep neural network models has become a standard practice for natural language processing tasks. From the performance on downstream tasks, we can know that the text representation constructed by these models contains linguistic information, but its encoding mode and extent are unclear. In this work, a structural probe is proposed to detect whether the vector representation produced by a deep neural network is embedded with a syntax tree. The probe is trained with the deep metric learning method, so that the distance between word vectors in the metric space it defines encodes the distance of words on the syntax tree, and the norm of word vectors encodes the depth of words on the syntax tree. The experiment results on ELMo and BERT show that the syntax tree is encoded in their parameters and the word representations they produce.

Keywords: deep metric learning, syntax tree probing, natural language processing, word representations

Procedia PDF Downloads 68

Authors: RAM PAL SINGH, VIKASH CHAUDHARY, MONIKA VERMA

Abstract:

In this paper, we proposed the implementation of optimization based Extreme Learning Machine (ELM) for watermarking of B-channel of color image in discrete wavelet transform (DWT) domain. ELM, a regularization algorithm, works based on generalized single-hidden-layer feed-forward neural networks (SLFNs). However, hidden layer parameters, generally called feature mapping in context of ELM need not to be tuned every time. This paper shows the embedding and extraction processes of watermark with the help of ELM and results are compared with already used machine learning models for watermarking.Here, a cover image is divide into suitable numbers of non-overlapping blocks of required size and DWT is applied to each block to be transformed in low frequency sub-band domain. Basically, ELM gives a unified leaning platform with a feature mapping, that is, mapping between hidden layer and output layer of SLFNs, is tried for watermark embedding and extraction purpose in a cover image. Although ELM has widespread application right from binary classification, multiclass classification to regression and function estimation etc. Unlike SVM based algorithm which achieve suboptimal solution with high computational complexity, ELM can provide better generalization performance results with very small complexity. Efficacy of optimization method based ELM algorithm is measured by using quantitative and qualitative parameters on a watermarked image even though image is subjected to different types of geometrical and conventional attacks.

Keywords: BER, DWT, extreme leaning machine (ELM), PSNR

Procedia PDF Downloads 311

Authors: Mabel Usunobun Olanipekun, Henry Ogbemudia Omoregbee

Abstract:

Anomalies such as water pipeline and hydraulic or petrochemical pipeline network leakages and bursts have significant implications for economic conditions and the environment. In order to ensure pipeline systems are reliable, they must be efficiently controlled. Wireless Sensor Networks (WSNs) have become a powerful network with critical infrastructure monitoring systems for water, oil and gas pipelines. The loss of water, oil and gas is inevitable and is strongly linked to financial costs and environmental problems, and its avoidance often leads to saving of economic resources. Substantial repair costs and the loss of precious natural resources are part of the financial impact of leaking pipes. Pipeline systems experts have implemented various methodologies in recent decades to identify and locate leakages in water, oil and gas supply networks. These methodologies include, among others, the use of acoustic sensors, measurements, abrupt statistical analysis etc. The issue of leak quantification is to estimate, given some observations about that network, the size and location of one or more leaks in a water pipeline network. In detecting background leakage, however, there is a greater uncertainty in using these methodologies since their output is not so reliable. In this work, we are presenting a scalable concept and simulation where a pressure-driven model (PDM) was used to determine water pipeline leakage in a system network. These pressure data were collected with the use of acoustic sensors located at various node points after a predetermined distance apart. We were able to determine with the use of correlation difference to determine the leakage point locally introduced at a predetermined point between two consecutive nodes, causing a substantial pressure difference between in a pipeline network. After de-noising the signal from the sensors at the nodes, we successfully obtained the exact point where we introduced the local leakage using the correlation difference model we developed.

Keywords: leakage detection, acoustic signals, pipeline network, correlation, wireless sensor networks (WSNs)

Procedia PDF Downloads 109

Authors: Victor Bodell, Lukas Ekstrom, Somayeh Aghanavesi

Abstract:

Fuel consumption (FC) is one of the key factors in determining expenses of operating a heavy-duty vehicle. A customer may therefore request an estimate of the FC of a desired vehicle. The modular design of heavy-duty vehicles allows their construction by specifying the building blocks, such as gear box, engine and chassis type. If the combination of building blocks is unprecedented, it is unfeasible to measure the FC, since this would first r equire the construction of the vehicle. This paper proposes a machine learning approach to predict FC. This study uses around 40,000 vehicles specific and o perational e nvironmental c onditions i nformation, such as road slopes and driver profiles. A ll v ehicles h ave d iesel engines and a mileage of more than 20,000 km. The data is used to investigate the accuracy of machine learning algorithms Linear regression (LR), K-nearest neighbor (KNN) and Artificial n eural n etworks (ANN) in predicting fuel consumption for heavy-duty vehicles. Performance of the algorithms is evaluated by reporting the prediction error on both simulated data and operational measurements. The performance of the algorithms is compared using nested cross-validation and statistical hypothesis testing. The statistical evaluation procedure finds that ANNs have the lowest prediction error compared to LR and KNN in estimating fuel consumption on both simulated and operational data. The models have a mean relative prediction error of 0.3% on simulated data, and 4.2% on operational data.

Keywords: artificial neural networks, fuel consumption, friedman test, machine learning, statistical hypothesis testing

Procedia PDF Downloads 178

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