Search results for: trained athletes

Authors: Amal Khalifa, Nicolas Vana Santos

Abstract:

Steganography has been known for centuries as an efficient approach for covert communication. Due to its popularity and ease of access, image steganography has attracted researchers to find secure techniques for hiding information within an innocent looking cover image. In this research, we propose a novel deep-learning approach to digital image steganography. The proposed method, DeepWaveletFusion, uses convolutional neural networks (CNN) to hide a secret image into a cover image of the same size. Two CNNs are trained back-to-back to merge the Discrete Wavelet Transform (DWT) of both colored images and eventually be able to blindly extract the hidden image. Based on two different image similarity metrics, a weighted gain function is used to guide the learning process and maximize the quality of the retrieved secret image and yet maintaining acceptable imperceptibility. Experimental results verified the high recoverability of DeepWaveletFusion which outperformed similar deep-learning-based methods.

Keywords: deep learning, steganography, image, discrete wavelet transform, fusion

Procedia PDF Downloads 91

Authors: Majid Arvan, Peyman Beygi, Sina Rokhsati

Abstract:

In-time detection of fire in buildings is of great importance. Employing intelligent methods in data processing in fire detection systems leads to a significant reduction of fire damage at lowest cost. In this paper, the raw data obtained from the fire detection sensor networks in buildings is processed by using intelligent methods based on neural networks and the likelihood of fire happening is predicted. In order to enhance the quality of system, the noise in the sensor data is reduced by analyzing wavelets and applying SVD technique. Meanwhile, the proposed neural network is trained using particle swarm optimization (PSO). In the simulation work, the data is collected from sensor network inside the room and applied to the proposed network. Then the outputs are compared with conventional MLP network. The simulation results represent the superiority of the proposed method over the conventional one.

Keywords: intelligent fire detection, neural network, particle swarm optimization, fire sensor network

Procedia PDF Downloads 380

Authors: Fuad M. Alkoot

Abstract:

Different experimental technologies such as microarray sequencing have been proposed to generate high-resolution genetic data, in order to understand the complex dynamic interactions between complex diseases and the biological system components of genes and gene products. However, the generated samples have a very large dimension reaching thousands. Therefore, hindering all attempts to design a classifier system that can identify diseases based on such data. Additionally, the high overlap in the class distributions makes the task more difficult. The data we experiment with is generated for the identification of autism. It includes 142 samples, which is small compared to the large dimension of the data. The classifier systems trained on this data yield very low classification rates that are almost equivalent to a guess. We aim at reducing the data dimension and improve it for classification. Here, we experiment with applying a multistage PCA on the genetic data to reduce its dimensionality. Results show a significant improvement in the classification rates which increases the possibility of building an automated system for autism detection.

Keywords: PCA, gene expression, dimensionality reduction, classification, autism

Procedia PDF Downloads 560

Authors: A. Hakimiyan, H. Namazi

Abstract:

Human behavior is defined as a range of behaviors exhibited by humans who are influenced by different internal or external sources. Human behavior is the subject of much research in different areas of psychology and neuroscience. Despite some advances in studies related to forecasting of human behavior, there are not many researches which consider the effect of the time delay between the presence of stimulus and the related human response. Analysis of EEG signal as a fractal time series is one of the major tools for studying the human behavior. In the other words, the human brain activity is reflected in his EEG signal. Artificial Neural Network has been proved useful in forecasting of different systems’ behavior especially in engineering areas. In this research, a time delay neural network is trained and tested in order to forecast the human EEG signal and subsequently human behavior. This neural network, by introducing a time delay, takes care of the lagging time between the occurrence of the stimulus and the rise of the subsequent action potential. The results of this study are useful not only for the fundamental understanding of human behavior forecasting, but shall be very useful in different areas of brain research such as seizure prediction.

Keywords: human behavior, EEG signal, time delay neural network, prediction, lagging time

Procedia PDF Downloads 663

Authors: Pavel Bulai, Taras Pitlik, Tatsiana Kulahava, Timofei Lipski

Abstract:

The method of Electrocardiography (ECG) interpretation for post-exercise recovery tracking was developed. Metabolic indices (aerobic and anaerobic) were designed using ECG-derived features. This study reports the associations between aerobic and anaerobic indices and classical parameters of the person’s physiological state, including blood biochemistry, glycogen concentration and VO2max changes. During the study 9 participants, healthy, physically active medium trained men and women, which trained 2-4 times per week for at least 9 weeks, fulfilled (i) ECG monitoring using Apple Watch Series 4 (AWS4); (ii) blood biochemical analysis; (iii) maximal oxygen consumption (VO2max) test, (iv) bioimpedance analysis (BIA). ECG signals from a single-lead wrist-wearable device were processed with detection of QRS-complex. Aerobic index (AI) was derived as the normalized slope of QR segment. Anaerobic index (ANI) was derived as the normalized slope of SJ segment. Biochemical parameters, glycogen content and VO2max were evaluated eight times within 3-60 hours after training. ECGs were recorded 5 times per day, plus before and after training, cycloergometry and BIA. The negative correlation between AI and blood markers of the muscles functional status including creatine phosphokinase (r=-0.238, p < 0.008), aspartate aminotransferase (r=-0.249, p < 0.004) and uric acid (r = -0.293, p<0.004) were observed. ANI was also correlated with creatine phosphokinase (r= -0.265, p < 0.003), aspartate aminotransferase (r = -0.292, p < 0.001), lactate dehydrogenase (LDH) (r = -0.190, p < 0.050). So, when the level of muscular enzymes increases during post-exercise fatigue, AI and ANI decrease. During recovery, the level of metabolites is restored, and metabolic indices rising is registered. It can be concluded that AI and ANI adequately reflect the physiology of the muscles during recovery. One of the markers of an athlete’s physiological state is the ratio between testosterone and cortisol (TCR). TCR provides a relative indication of anabolic-catabolic balance and is considered to be more sensitive to training stress than measuring testosterone and cortisol separately. AI shows a strong negative correlation with TCR (r=-0.437, p < 0.001) and correctly represents post-exercise physiology. In order to reveal the relation between the ECG-derived metabolic indices and the state of the cardiorespiratory system, direct measurements of VO2max were carried out at various time points after training sessions. The negative correlation between AI and VO2max (r = -0.342, p < 0.001) was obtained. These data testifying VO2max rising during fatigue are controversial. However, some studies have revealed increased stroke volume after training, that agrees with findings. It is important to note that post-exercise increase in VO2max does not mean an athlete’s readiness for the next training session, because the recovery of the cardiovascular system occurs over a substantially longer period. Negative correlations registered for ANI with glycogen (r = -0.303, p < 0.001), albumin (r = -0.205, p < 0.021) and creatinine (r = -0.268, p < 0.002) reflect the dehydration status of participants after training. Correlations between designed metabolic indices and physiological parameters revealed in this study can be considered as the sufficient evidence to use these indices for assessing the state of person’s aerobic and anaerobic metabolic systems after training during fatigue, recovery and supercompensation.

Keywords: aerobic index, anaerobic index, electrocardiography, supercompensation

Procedia PDF Downloads 115

Authors: Simon Strande Henriksen

Abstract:

When a city and country decide to undertake an Olympic Games, they do so with the notion that hosting the Olympics will provide direct financial benefits to the city, country, and national companies. Like many activities, the Olympic Games tend to be more popular when it is warm, and the athletes are known, and therefore this paper will only focus on the two latest Olympic Summer Games. Cities and countries continue to invest billions of dollars in infrastructure to secure the role of being Olympic hosts. The multiple investments expect to provide both economic growth and a lasting legacy for the citizens. This study aims to determine whether host country companies experience superior economic impact from the Olympics. Building on existing work within the Olympic field of research, it asks: Do companies in host countries of the Olympic Summer Games experience a superior increase in operating revenue and return on assets compared to other comparable countries? In this context, comparable countries are the two candidates following the host city in the bidding procedure. Based on methods used by scholars, a panel data regression was conducted on revenue growth rate and return on assets, to determine if host country companies see a positive relation with hosting the Olympic Games. Combined with an analysis of motivation behind hosting the Olympics, the regression showed no significant positive relations across all analyses, besides in one instance. Indications of a relationship between company performance and economic motivation were found to be present. With the results indicating a limited effect on company growth, it is recommended that prospective host cities and countries carefully consider possible implications the role of being an Olympic host might have on national companies.

Keywords: cross-country analysis, mega-event, multiple regression, quantitative analysis

Procedia PDF Downloads 141

Authors: Yu Liang, Wei Wei

Abstract:

Spatial representation of numbers and numerical magnitude are important for preschoolers’ mathematical ability. Mental number line, a typical index to measure numbers spatial representation, and numerical comparison are both related to arithmetic obviously. However, they seem to rely on different mechanisms and probably influence arithmetic through different mechanisms. In line with this idea, preschool children were trained with two tasks to investigate which one is more important for approximate arithmetic. The training of numerical processing and number line estimation were proved to be effective. They both improved the ability of approximate arithmetic. When the difficulty of approximate arithmetic was taken into account, the performance in number line training group was not significantly different among three levels. However, two harder levels achieved significance in numerical comparison training group. Thus, comparing spatial representation ability, symbolic approximation arithmetic relies more on numerical magnitude. Educational implications of the study were discussed.

Keywords: approximate arithmetic, mental number line, numerical magnitude, preschooler

Procedia PDF Downloads 252

Authors: Zainuddin Zakaria, Sun Woh Lye

Abstract:

Despite the continuous advancements in automated conflict resolution tools, there is still a low rate of adoption of automation from Air Traffic Control Officers (ATCOs). Trust or acceptance in these tools and conformance to the individual ATCO preferences in strategy execution for conflict resolution are two key factors that impact their use. This paper proposes a methodology to unearth and classify ATCO conflict resolution strategies from simulator data of trained and qualified ATCOs. The methodology involves the extraction of ATCO executive control actions and the establishment of a system of strategy resolution classification based on ATCO radar commands and prevailing flight parameters in deconflicting a pair of aircraft. Six main strategies used to handle various categories of conflict were identified and discussed. It was found that ATCOs were about twice more likely to choose only vertical maneuvers in conflict resolution compared to horizontal maneuvers or a combination of both vertical and horizontal maneuvers.

Keywords: air traffic control strategies, conflict resolution, simulator data, strategy classification system

Procedia PDF Downloads 148

Authors: Jay-R M. Mendoza

Abstract:

In almost all mathematics classrooms, students demonstrated discrepancies in their knowledge, skills, and understanding. OECD reports predicted that this continued to aggravate as not all teachers were sufficiently trained to handle this concentration. In response, the paper explored the potential of reSolve’s professional learning module 3 (PLM3) as an affordable and accessible professional development (PD) resource. Participants’ hands-on experience and exposure to PLM3 were audio recorded. After it was transcribed and examined and their work samples were analysed, there were four issues emerged: (1) criticality of conducting preliminary data collections and increasing the validity of inferences about what students can and cannot do by addressing the probabilistic nature of their performance; (2) criticality of the conclusion: a > b and/or (a-b) ∈ Z⁺ among students’ algebraic reasoning; (3) enabling and extending prompts provided by reSolve were found useful; and (4) dynamic adaptation of reSolve PLM3 through developing transferable skills and collaboration among teachers. PLM3 provided valuable insights on assessment, teaching, and planning to include all students in mathematics experiences.

Keywords: algebraic reasoning, building teacher capacity, including all students in mathematics experiences, professional development

Procedia PDF Downloads 124

Authors: Eisa M. Al-Balhan, Mamdouh M. Soliman

Abstract:

This study introduced learning styles techniques into mathematics teaching to improve mathematics achievement and motivation among Kuwaiti fourth- and fifth-grade low achievers. The study consisted of two groups. The control group (N = 212) received traditional math tutoring based on a textbook and the tutor’s knowledge of math. The experimental group (N = 209) received math tutoring from instructors trained in the Learning Style™ approach. Three instruments were used: Motivation Scale towards Mathematics; Achievement in Mathematics Test; and the manual of learning style approach indicating the individual’s preferred learning style: AKV, AVK, KAV, KVA, VAK, or VKA. The participating teachers taught to the detected learning style of each student or group. The findings show significant improvement in achievement and motivation towards mathematics in the experimental group. The outcome offers information to variables affecting achievement and motivation towards mathematics and demonstrates the leading role of Kuwait in education within the region.

Keywords: elementary school, learning style, math low achievers, SmartWired™, math instruction, motivation

Procedia PDF Downloads 111

Authors: Nathaniel C. Villanueva, Ian K. H. Chun, Alyssa S. Fujiwara, Emily R. Leibovitch, Brennan E. Yamamoto, Loren G. Yamamoto

Abstract:

Introduction: In previous studies of high school sports, over 15% of concussions were attributed to contact with the playing surface. While artificial turf fields are increasing in popularity due to lower maintenance costs, artificial turf has been associated with more ankle and knee injuries, with inconclusive data on concussions. In this study, natural grass and artificial football fields were compared in terms of deceleration on fall impact. Methods: Accelerometers were placed on the forehead, apex of the head, and right ear of a Century Body Opponent Bag (BOB) manikin. A Riddell HITS football helmet was secured onto the head of the manikin over the accelerometers. This manikin was dropped onto natural grass (n = 10) and artificial turf (n = 9) high school football fields. The manikin was dropped from a stationary position at a height of 60 cm onto its front, back, and left side. Each of these drops was conducted 10 times at the 40-yard line, 20-yard line, and endzone. The net deceleration on impact was calculated as a net vector from each of the three accelerometers’ x, y, and z vectors from the three different locations on the manikin’s head (9 vector measurements per drop). Results: Mean values for the multiple drops were calculated for each accelerometer and drop type for each field. All accelerometers in forward and backward falls and one accelerometer in side falls showed significantly greater impact force on synthetic turf compared to the natural grass surfaces. Conclusion: Impact force was higher on synthetic fields for all drop types for at least one of the accelerometer locations. These findings suggest that concussion risk might be higher for athletes playing on artificial turf fields.

Keywords: concussion, football, biomechanics, sports

Procedia PDF Downloads 158

Authors: Delphine Ribes Lemay, Valentine Bernasconi, André Andrade, Lara DéFayes, Mathieu Salzmann, FréDéRic Kaplan, Nicolas Henchoz

Abstract:

Digitization and preservation of large heritage induce high maintenance costs to keep up with the technical standards and ensure sustainable access. Creating impactful usage is instrumental to justify the resources for long-term preservation. The Museum für Gestaltung of Zurich holds one of the biggest poster collections of the world from which 52’000 were digitised. In the process of building a digital installation to valorize the collection, one objective was to develop an algorithm capable of predicting the next poster to show according to the ones already displayed. The work presented here describes the steps to build an algorithm able to automatically create sequences of posters reflecting associations performed by curator and professional designers. The exposed challenge finds similarities with the domain of song playlist algorithms. Recently, artificial intelligence techniques and more specifically, deep-learning algorithms have been used to facilitate their generations. Promising results were found thanks to Recurrent Neural Networks (RNN) trained on manually generated playlist and paired with clusters of extracted features from songs. We used the same principles to create the proposed algorithm but applied to a challenging medium, posters. First, a convolutional autoencoder was trained to extract features of the posters. The 52’000 digital posters were used as a training set. Poster features were then clustered. Next, an RNN learned to predict the next cluster according to the previous ones. RNN training set was composed of poster sequences extracted from a collection of books from the Gestaltung Museum of Zurich dedicated to displaying posters. Finally, within the predicted cluster, the poster with the best proximity compared to the previous poster is selected. The mean square distance between features of posters was used to compute the proximity. To validate the predictive model, we compared sequences of 15 posters produced by our model to randomly and manually generated sequences. Manual sequences were created by a professional graphic designer. We asked 21 participants working as professional graphic designers to sort the sequences from the one with the strongest graphic line to the one with the weakest and to motivate their answer with a short description. The sequences produced by the designer were ranked first 60%, second 25% and third 15% of the time. The sequences produced by our predictive model were ranked first 25%, second 45% and third 30% of the time. The sequences produced randomly were ranked first 15%, second 29%, and third 55% of the time. Compared to designer sequences, and as reported by participants, model and random sequences lacked thematic continuity. According to the results, the proposed model is able to generate better poster sequencing compared to random sampling. Eventually, our algorithm is sometimes able to outperform a professional designer. As a next step, the proposed algorithm should include a possibility to create sequences according to a selected theme. To conclude, this work shows the potentiality of artificial intelligence techniques to learn from existing content and provide a tool to curate large sets of data, with a permanent renewal of the presented content.

Keywords: Artificial Intelligence, Digital Humanities, serendipity, design research

Procedia PDF Downloads 184

Authors: Ashkan Zakaryazad, Ekrem Duman

Abstract:

A typical classification technique ranks the instances in a data set according to the likelihood of belonging to one (positive) class. A credit card (CC) fraud detection model ranks the transactions in terms of probability of being fraud. In fact, this approach is often criticized, because firms do not care about fraud probability but about the profitability or costliness of detecting a fraudulent transaction. The key contribution in this study is to focus on the profit maximization in the model building step. The artificial neural network proposed in this study works based on profit maximization instead of minimizing the error of prediction. Moreover, some studies have shown that the back propagation algorithm, similar to other gradient–based algorithms, usually gets trapped in local optima and swarm-based algorithms are more successful in this respect. In this study, we train our profit maximization ANN using the Migrating Birds optimization (MBO) which is introduced to literature recently.

Keywords: neural network, profit-based neural network, sum of squared errors (SSE), MBO, gradient descent

Procedia PDF Downloads 475

Authors: Alexander Hudson, Shaogang Gong

Abstract:

Structure determination is key to understanding protein function at a molecular level. Whilst significant advances have been made in predicting structure and function from amino acid sequence, researchers must still rely on expensive, time-consuming analytical methods to visualise detailed protein conformation. In this study, we demonstrate that it is possible to make accurate (≥80%) predictions of protein class and architecture from structures determined at low (>3A) resolution, using a deep convolutional neural network trained on high-resolution (≤3A) structures represented as 2D matrices. Thus, we provide proof of concept for high-speed, low-cost protein structure classification at low resolution, and a basis for extension to prediction of function. We investigate the impact of the input representation on classification performance, showing that side-chain information may not be necessary for fine-grained structure predictions. Finally, we confirm that high resolution, low-resolution and NMR-determined structures inhabit a common feature space, and thus provide a theoretical foundation for boosting with single-image super-resolution.

Keywords: transfer learning, protein distance maps, protein structure classification, neural networks

Procedia PDF Downloads 136

Authors: Magdy Abouzeid

Abstract:

Introduction: Physical activity has been shown to have a positive effect on bone mineral density (BMD) and bone mineral content (BMC) among children, adolescents, and adults. Sports characterized by little or moderate weight bearing or impact have a low osteogenic effect. However, the action of such sports on bone turnover remains unclear. Swimming, as a non-weight-bearing sport, has been considered to be insignificant in the maintenance of bone mass. Purpose: To examine this issue we measured (BMD) and(BMC) of the lumbar spine, proximal femur via dual energy x-ray absorptiometry in the group of elite male swimmers, and determine the effect of swimming training on bone health and compared the results with matched controls group in age, body weight and height. Materials and Methods: Twenty-five male swimmers (age 20.7+/-0.8 years) training for 12-15 hours/week; and the controls group consisted of 25 non-active male (age 21.3 +/-1.3 years) were studied BMD and BMC of lumbar spine, femur were assessed via (DXA) absorptiometry. Results: There was significant difference between swimmers and control group in BMD and BMC, BMD of Swimmers was significantly greater than controls at all sites. The lumbar spine (1, 08 +/-0.202 vs., 0717+0.57 gxcm (-2), right proximal femur (1, 02 +/-, 044 vs., 771+/-, 027 gxcm (-2), and left proximal femur (1.374+/-0.212 vs. 1.01 +/-0.141 gxcm (-2). Swimmers were significantly taller, and had greater BMC and BMD compared to the controls group (P<0.001). Conclusions: These results suggest that swimming training may be beneficial in the prevention or therapy of OSTEOPENIA, and may lead to increased (BMD) and (BMC) for male swimmers. Swimming may be an effective non-pharmacological intervention for the adults and adolescent. Further research with younger athletes of another type of aquatics sport is warranted to better identify the periods of BMD development during which Aquatics sport has the greatest impact on bone health.

Keywords: bone mineral density, lumbar spine, femur, swimming, DXA absorptiometry

Procedia PDF Downloads 322

Authors: Jinan Fiaidhi, Sabah Mohammed

Abstract:

Cross-domain learning in the medical field is a research challenge as many conditions, like in oncology imaging, use different imaging modalities. Moreover, in most of the medical learning applications, the sample training size is relatively small. Although few-shot learning (FSL) through the use of a Siamese neural network was able to be trained on a small sample with remarkable accuracy, FSL fails to be effective for use in multiple domains as their convolution weights are set for task-specific applications. In this paper, we are addressing this problem by enabling FSL to possess the ability to shift across domains by designing a two-layer FSL network that can learn individually from each domain and produce a shared features map with extra modulation to be used at the second layer that can recognize important targets from mix domains. Our initial experimentations based on mixed medical datasets like the Medical-MNIST reveal promising results. We aim to continue this research to perform full-scale analytics for testing our cross-domain FSL learning.

Keywords: Siamese neural network, few-shot learning, meta-learning, metric-based learning, thick data transformation and analytics

Procedia PDF Downloads 56

Authors: Tathiana Yesenia Coy Mondragón, Jossitt William Vargas Cruz, Cristian Leonardo Gutiérrez Gómez

Abstract:

The study of cosmic rays is based on two fundamental pillars: the detection of secondary cosmic rays on the Earth's surface and the detection of the source and origin of the cascade. In addition, the constant flow of RC generates a lot of interest for study due to the incidence of various natural phenomena, which makes it relevant to characterize their incidence parameters to determine their effect not only at subsoil or terrestrial surface levels but also throughout the atmosphere. To determine the physical parameters of the primary cosmic ray, the implementation of robust algorithms capable of reconstructing the cascade from the measured values is required, with a high level of reliability. Therefore, it is proposed to build a machine learning system that will be fed from the cosmic ray simulations in CORSIKA at different energies that lie in a range [10⁹-10¹²] eV. in order to generate a trained particle and pattern recognition system to obtain greater efficiency when inferring the nature of the origin of the cascade for EAS in the atmosphere considering atmospheric models.

Keywords: CORSIKA, cosmic rays, eas, Colombia

Procedia PDF Downloads 81

Authors: B. R. Campomanes-Alvarez, P. Quiros, B. Fernandez

Abstract:

Automatic Speech Recognition (ASR) is a machine-based process of decoding and transcribing oral speech. A typical ASR system receives acoustic input from a speaker or an audio file, analyzes it using algorithms, and produces an output in the form of a text. Some speech recognition systems use Hidden Markov Models (HMMs) to deal with the temporal variability of speech and Gaussian Mixture Models (GMMs) to determine how well each state of each HMM fits a short window of frames of coefficients that represents the acoustic input. Another way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition systems. Acoustic models for state-of-the-art ASR systems are usually training on massive amounts of data. However, audio files with their corresponding transcriptions can be difficult to obtain, especially in the Spanish language. Hence, in the case of these low-resource scenarios, building an ASR model is considered as a complex task due to the lack of labeled data, resulting in an under-trained system. Semi-supervised learning approaches arise as necessary tasks given the high cost of transcribing audio data. The main goal of this proposal is to develop a procedure based on acoustic semi-supervised learning for Spanish ASR systems by using DNNs. This semi-supervised learning approach consists of: (a) Training a seed ASR model with a DNN using a set of audios and their respective transcriptions. A DNN with a one-hidden-layer network was initialized; increasing the number of hidden layers in training, to a five. A refinement, which consisted of the weight matrix plus bias term and a Stochastic Gradient Descent (SGD) training were also performed. The objective function was the cross-entropy criterion. (b) Decoding/testing a set of unlabeled data with the obtained seed model. (c) Selecting a suitable subset of the validated data to retrain the seed model, thereby improving its performance on the target test set. To choose the most precise transcriptions, three confidence scores or metrics, regarding the lattice concept (based on the graph cost, the acoustic cost and a combination of both), was performed as selection technique. The performance of the ASR system will be calculated by means of the Word Error Rate (WER). The test dataset was renewed in order to extract the new transcriptions added to the training dataset. Some experiments were carried out in order to select the best ASR results. A comparison between a GMM-based model without retraining and the DNN proposed system was also made under the same conditions. Results showed that the semi-supervised ASR-model based on DNNs outperformed the GMM-model, in terms of WER, in all tested cases. The best result obtained an improvement of 6% relative WER. Hence, these promising results suggest that the proposed technique could be suitable for building ASR models in low-resource environments.

Keywords: automatic speech recognition, deep neural networks, machine learning, semi-supervised learning

Procedia PDF Downloads 339

Authors: Jose Ramon Sanabria Navarro, Yahilina Silveira Perez

Abstract:

The Pan American Games are one of the best regional sports integration events for the Americas, thousands of athletes from different countries are integrated based on obtaining satisfactory sports results and bringing the glory of the national identity of the sport to their country. There are countries that despite the small number of inhabitants have obtained very satisfactory results such as Cuba. Objective: To analyze the impact of Cuba's sports results in the Pan American Games. The methodology was based on the postulates of the materialist dialectic since the investigated reality was studied from historicity, systematicity and in its systemic character. The population and sample consisted of 41 countries of the American continent, and the 15 events carried out to date were analyzed. The impact of Cuba is very relevant because it is the country that has the second place by country in terms of number of medals, is among the first in terms of medals per inhabitants and in general sense of all the indicators treated assumes the fourth place integral. What is the current status of Cuba's sports results in Pan American Games? Having as a general objective, analyze the impact of Cuba's sports results in the Pan American Games. The hypotheses that will lead this research have the following methodological and interaction order: H1: Cuba's performance in Pan American Sports Games positively impacts the amount of medals obtained. H2: The amount of medals from Cuba in Pan American Sports Games positively impacts the general podium for countries of these regional events. H3: The amount of medals obtained by Cuba in Pan American Sports Games positively impacts the number of inhabitants. H4: The amount of medals obtained by Cuba positively impacts the overall result of the countries. H5: Cuba's performance in the Panamerican Sports Games positively impacts the overall results of these regional events. In general, it is possible to demonstrate the impact of Cuba's sports performance in Pan American Games and the organizational sports structure that has allowed the country to obtain them is evidenced.

Keywords: Cuba, history of sport, sports games, regional events, sport

Procedia PDF Downloads 128

Authors: Neeta Kumari, Gopal Pathak

Abstract:

Artificial neural network has been proved to be an efficient tool for non-parametric modeling of data in various applications where output is non-linearly associated with input. It is a preferred tool for many predictive data mining applications because of its power , flexibility, and ease of use. A standard feed forward networks (FFN) is used to predict the groundwater fluoride content. The ANN model is trained using back propagated algorithm, Tansig and Logsig activation function having varying number of neurons. The models are evaluated on the basis of statistical performance criteria like Root Mean Squarred Error (RMSE) and Regression coefficient (R2), bias (mean error), Coefficient of variation (CV), Nash-Sutcliffe efficiency (NSE), and the index of agreement (IOA). The results of the study indicate that Artificial neural network (ANN) can be used for groundwater fluoride prediction in the limited data situation in the hard rock region like western parts of Jharkhand with sufficiently good accuracy.

Keywords: Artificial neural network (ANN), FFN (Feed-forward network), backpropagation algorithm, Levenberg-Marquardt algorithm, groundwater fluoride contamination

Procedia PDF Downloads 550

Authors: A. A. Adekunle, A. M. Omoare, W. O. Oyediran

Abstract:

This study was carried out to assess rural women’s skill acquisition in the processing of locust bean in Ipokia Local Government Area of Ogun State, Nigeria. Simple random sampling technique was used to select 90 women locust bean processors for this study. Data were analyzed with descriptive statistics and Pearson Product Moment Correlation. The result showed that the mean age of respondents was 40.72 years. Most (70.00%) of the respondents were married. The mean processing experience was 8.63 years. 93.30% of the respondents relied on information from fellow locust beans processors and friends. All (100%) the respondents did not acquire improved processing skill through trainings and workshops. It can be concluded that the rural women’s skill acquisition on modernized processing techniques was generally low. It is hereby recommend that the rural women processors should be trained by extension service providers through series of workshops and seminars on improved processing techniques.

Keywords: locust bean, processing, skill acquisition, rural women

Procedia PDF Downloads 461

Authors: Peng Wang, Tigang Jiang

Abstract:

The computer based lip reading method of different languages cannot be universal. At present, for the research of Chinese lip reading, whether the work on data sets or recognition algorithms, is far from mature. In this paper, we study the Chinese lipreading method based on machine learning, and propose a Chinese Sentence-level lip-reading network (CNLipNet) model which consists of spatio-temporal convolutional neural network(CNN), recurrent neural network(RNN) and Connectionist Temporal Classification (CTC) loss function. This model can map variable-length sequence of video frames to Chinese Pinyin sequence and is trained end-to-end. More over, We create CNLRS, a Chinese Lipreading Dataset, which contains 5948 samples and can be shared through github. The evaluation of CNLipNet on this dataset yielded a 41% word correct rate and a 70.6% character correct rate. This evaluation result is far superior to the professional human lip readers, indicating that CNLipNet performs well in lipreading.

Keywords: lipreading, machine learning, spatio-temporal, convolutional neural network, recurrent neural network

Procedia PDF Downloads 128

Authors: Go Woon Kwak, Siyoung Jun, Soyun Maeng, Haeyoung Lee

Abstract:

Entry is an education platform used in South Korea, created to help students learn to program, in which they can learn to code while playing. Using the online version of the entry, teachers can easily assign programming homework to the student and the students can make programs simply by linking programming blocks. However, the programs may be made by others, so that the authors of the programs should be identified. In this paper, as the first step toward author identification of entry programs, we present an artificial neural network based classification approach to identify genders of authors of a program written in an entry. A neural network has been trained from labeled training data that we have collected. Our result in progress, although preliminary, shows that the proposed approach could be feasible to be applied to the online version of entry for gender identification of authors. As future work, we will first use a machine learning technique for age identification of entry programs, which would be the second step toward the author identification.

Keywords: artificial intelligence, author identification, deep neural network, gender identification, machine learning

Procedia PDF Downloads 324

Authors: Nurwina Anuar, Aswad Anuar

Abstract:

Imagery is a popular mental technique used by athletes and coaches to improve learning and performance. It has been widely investigated and beneficial in the sports context. However, the imagery application in equestrian sport has been understudied. Thus, the effectiveness of imagery should encompass the application in the equestrian sport to ensure its application covert all sports. Unlike most sports (e.g., football, badminton, tennis, ski) which are both mental and physical are dependent solely upon human decision and response, equestrian sports involves the interaction of human-horse collaboration to success in the equestrian tasks. This study aims to investigate the effect of PETTLEP imagery on equestrian jumping tasks, motivation and imagery ability. It was hypothesized that the use of PETTLEP imagery intervention will significantly increase in the skill equestrian jumping tasks. It was also hypothesized that riders’ imagery ability and motivation will increase across phases. The participants were skilled riders with less to no imagery experience. A single-subject ABA design was employed. The study was occurred over five week’s period at Universiti Teknologi Malaysia Equestrian Park. Imagery ability was measured using the Sport Imagery Assessment Questionnaires (SIAQ), the motivational measured based on the Motivational imagery ability measure for Sport (MIAMS). The effectiveness of the PETTLEP imagery intervention on show jumping tasks were evaluated by the professional equine rider on the observational scale. Results demonstrated the improvement on all equestrian jumping tasks for the most participants from baseline to intervention. Result shows the improvement on imagery ability and participants’ motivations after the PETTLEP imagery intervention. Implication of the present study include underlining the impact of PETTLEP imagery on equestrian jumping tasks. The result extends the previous research on the effectiveness of PETTLEP imagery in the sports context that involves interaction and collaboration between human and horse.

Keywords: PETTLEP imagery, imagery ability, equestrian, equestrian jumping tasks

Procedia PDF Downloads 202

Authors: Lucie Hayashi

Abstract:

This paper offers a view on a different approach towards a dancer´s career in two very dissimilar countries: on one hand Japan, an economic predator at the end of last century, but suffering under economic crisis from the beginning of the new century; and the Czech Republic, a post-communist country, caught up in capitalist fever from the 1990s on the other. The government’s approach towards culture and dance in these two countries not only has a different history and nature, but also presents a different take on the ideal future development in its respective dance scenes. The level of support from the state budget echoes in all the fields of a professional dance career, dance art and the education of the public towards dance. The message of the statistic data is clear: the production of an enormous number of well trained and expensively educated dancers with no jobs for them in Japan, and a lack of good dancers ready to fill state supported theatre companies in the Czech Republic (that gladly employs Japanese dancers). The paradigm leaves a big exclamation mark on the huge influence the policy has on dance in society, and a question mark on the ideal situation.

Keywords: culture policy, dance, education, employment, Czech Republic, Japan

Procedia PDF Downloads 164

Authors: Chico Horacio Jose Sambo

Abstract:

Recently neural network has gained popularity when come to solve complex nonlinear problems. Permeability is one of fundamental reservoir characteristics system that are anisotropic distributed and non-linear manner. For this reason, permeability prediction from well log data is well suited by using neural networks and other computer-based techniques. The main goal of this paper is to predict reservoir permeability from well logs data by using neural network approach. A multi-layered perceptron trained by back propagation algorithm was used to build the predictive model. The performance of the model on net results was measured by correlation coefficient. The correlation coefficient from testing, training, validation and all data sets was evaluated. The results show that neural network was capable of reproducing permeability with accuracy in all cases, so that the calculated correlation coefficients for training, testing and validation permeability were 0.96273, 0.89991 and 0.87858, respectively. The generalization of the results to other field can be made after examining new data, and a regional study might be possible to study reservoir properties with cheap and very fast constructed models.

Keywords: neural network, permeability, multilayer perceptron, well log

Procedia PDF Downloads 403

Authors: Azadeh Bashiri

Abstract:

Introduction: Lack of proper diagnosis and inadequate treatment of asthma leads to physical and financial complications. This study aimed to use data mining techniques and creating a neural network intelligent system for diagnosis of asthma. Methods: The study population is the patients who had visited one of the Lung Clinics in Tehran. Data were analyzed using the SPSS statistical tool and the chi-square Pearson's coefficient was the basis of decision making for data ranking. The considered neural network is trained using back propagation learning technique. Results: According to the analysis performed by means of SPSS to select the top factors, 13 effective factors were selected, in different performances, data was mixed in various forms, so the different models were made for training the data and testing networks and in all different modes, the network was able to predict correctly 100% of all cases. Conclusion: Using data mining methods before the design structure of system, aimed to reduce the data dimension and the optimum choice of the data, will lead to a more accurate system. Therefore, considering the data mining approaches due to the nature of medical data is necessary.

Keywords: asthma, data mining, Artificial Neural Network, intelligent system

Procedia PDF Downloads 273

Authors: Olga Lasaga, Carmen Parra

Abstract:

Within the EU 'New Skills for New Jobs' initiative, the art and music sector is considered one of the most vulnerable. Its graduates are faced with the threat of the dole or of not finding work in the sector in which they trained. In this regard, an increasing number of students are graduating every year from European Conservatories and Fine Arts Centres, while the number of job opportunities in this sector has stagnated or decreased. Moreover, the traditional teaching of these institutes does not favour the acquisition of basic skills, such as team building, entrepreneurship, marketing, website design and the design of events, which are among the most important facets of project management and are precisely those aspects that are often most related to the improvement of employability in the art world. To remedy this situation, the results of the European Erasmus+ OMEGA project (Opening More Employment Gates for Art and Music Students) are presented. The OMEGA project aims to increase the employability of art and music students by equipping them with additional skills needed for the search for work. As a result of this project, a manual has been created, a pilot course has been designed and taught, and a comparative study has been conducted on the state of play of the participating countries.

Keywords: artists, employability, entrepreneurship, musicians, skills

Procedia PDF Downloads 244

Authors: Sunghoon Park, Hank Kim, Jinwon An, Sungzoon Cho

Abstract:

Internet of Things allows one to collect data from facilities which are then used to monitor them and even predict malfunctions in advance. Conventional quality control methods focus on setting a normal range on a sensor value defined between a lower control limit and an upper control limit, and declaring as an anomaly anything falling outside it. However, interactions among sensor values are ignored, thus leading to suboptimal performance. We propose a multivariate approach which takes into account many sensor values at the same time. In particular Gaussian Mixture Model is used which is trained to maximize likelihood value using Expectation-Maximization algorithm. The number of Gaussian component distributions is determined by Bayesian Information Criterion. The negative Log likelihood value is used as an anomaly score. The actual usage scenario goes like a following. For each instance of sensor values from a facility, an anomaly score is computed. If it is larger than a threshold, an alarm will go off and a human expert intervenes and checks the system. A real world data from Building energy system was used to test the model.

Keywords: facility anomaly detection, gaussian mixture model, anomaly score, expectation maximization algorithm

Procedia PDF Downloads 272

Authors: Zhaoxin Luo, Michael Zhu

Abstract:

In natural languages, there are always complex semantic hierarchies. Obtaining the feature representation based on these complex semantic hierarchies becomes the key to the success of the model. Several RNN models have recently been proposed to use latent indicators to obtain the hierarchical structure of documents. However, the model that only uses a single-layer latent indicator cannot achieve the true hierarchical structure of the language, especially a complex language like Chinese. In this paper, we propose a deep layered model that stacks arbitrarily many RNN layers equipped with latent indicators. After using EM and training it hierarchically, our model solves the computational problem of stacking RNN layers and makes it possible to stack arbitrarily many RNN layers. Our deep hierarchical model not only achieves comparable results to large pre-trained models on the Chinese short text classification problem but also achieves state of art results on the Chinese long text classification problem.

Keywords: nature language processing, recurrent neural network, hierarchical structure, document classification, Chinese

Procedia PDF Downloads 68