Search results for: hidden Markov model toolkit (HTK)

Authors: Yaowei Hu, Yongkai Wu, Lu Zhang, Xintao Wu

Abstract:

Causal inference from observational data is receiving wide applications in many fields. However, unidentifiable situations, where causal effects cannot be uniquely computed from observational data, pose critical barriers to applying causal inference to complicated real applications. In this paper, we develop a bounding method for estimating the average causal effect (ACE) under unidentifiable situations due to hidden confounders. We propose to parameterize the unknown exogenous random variables and structural equations of a causal model using neural networks and implicit generative models. Then, with an adversarial learning framework, we search the parameter space to explicitly traverse causal models that agree with the given observational distribution and find those that minimize or maximize the ACE to obtain its lower and upper bounds. The proposed method does not make any assumption about the data generating process and the type of the variables. Experiments using both synthetic and real-world datasets show the effectiveness of the method.

Keywords: average causal effect, hidden confounding, bound estimation, generative adversarial learning

Procedia PDF Downloads 157

Authors: Ebert Brea

Abstract:

We propose a novel direct search algorithm for identifying at least a local minimum of mixed integer nonlinear unconstrained optimization problems. The Mixed Integer Randomized Pattern Search Algorithm (MIRPSA), so-called by the author, is based on a randomized pattern search, which is modified by the MIRPSA for finding at least a local minimum of our problem. The MIRPSA has two main operations over the randomized pattern search: moving operation and shrinking operation. Each operation is carried out by the algorithm when a set of conditions is held. The convergence properties of the MIRPSA is analyzed using a Markov chain approach, which is represented by an infinite countable set of state space λ, where each state d(q) is defined by a measure of the qth randomized pattern search Hq, for all q in N. According to the algorithm, when a moving operation is carried out on the qth randomized pattern search Hq, the MIRPSA holds its state. Meanwhile, if the MIRPSA carries out a shrinking operation over the qth randomized pattern search Hq, the algorithm will visit the next state, this is, a shrinking operation at the qth state causes a changing of the qth state into (q+1)th state. It is worthwhile pointing out that the MIRPSA never goes back to any visited states because the MIRPSA only visits any qth by shrinking operations. In this article, we describe the MIRPSA for mixed integer nonlinear unconstrained optimization problems for doing a deep study of its convergence properties using Markov chain viewpoint. We herein include a low dimension case for showing more details of the MIRPSA, when the algorithm is used for identifying the minimum of a mixed integer quadratic function. Besides, numerical examples are also shown in order to measure the performance of the MIRPSA.

Keywords: direct search, mixed integer optimization, random search, convergence, Markov chain

Procedia PDF Downloads 438

Authors: B. L. Ho, L. Shi, D. F. Wang, V. C. T. Mok

Abstract:

The brain’s functional connectivity while temporally non-stationary does express consistency at a macro spatial level. The study of stable resting state connectivity patterns hence provides opportunities for identification of diseases if such stability is severely perturbed. A mathematical model replicating the brain’s spatial connections will be useful for understanding brain’s representative geometry and complements the empirical model where it falls short. Empirical computations tend to involve large matrices and become infeasible with fine parcellation. However, the proposed analytical model has no such computational problems. To improve replicability, 92 subject data are obtained from two open sources. The proposed methodology, inspired by financial theory, uses multivariate regression to find relationships of every cortical region of interest (ROI) with some pre-identified hubs. These hubs acted as representatives for the entire cortical surface. A variance-covariance framework of all ROIs is then built based on these relationships to link up all the ROIs. The result is a high level of match between model and empirical correlations in the range of 0.59 to 0.66 after adjusting for sample size; an increase of almost forty percent. More significantly, the model framework provides an intuitive way to delineate between systemic drivers and idiosyncratic noise while reducing dimensions by more than 30 folds, hence, providing a way to conduct attribution analysis. Due to its analytical nature and simple structure, the model is useful as a standalone toolkit for network dependency analysis or as a module for other mathematical models.

Keywords: functional magnetic resonance imaging, multivariate regression, network hubs, resting state functional connectivity

Procedia PDF Downloads 130

Authors: Pubudu K. Hitigala Kaluarachchilage, Champike Attanayake, Sasith Rajasooriya, Chris P. Tsokos

Abstract:

Operating system (OS) security is a key component of computer security. Assessing and improving OSs strength to resist against vulnerabilities and attacks is a mandatory requirement given the rate of new vulnerabilities discovered and attacks occurring. Frequency and the number of different kinds of vulnerabilities found in an OS can be considered an index of its information security level. In the present study five mostly used OSs, Microsoft Windows (windows 7, windows 8 and windows 10), Apple’s Mac and Linux are assessed for their discovered vulnerabilities and the risk associated with each. Each discovered and reported vulnerability has an exploitability score assigned in CVSS score of the national vulnerability database. In this study the risk from vulnerabilities in each of the five Operating Systems is compared. Risk Indexes used are developed based on the Markov model to evaluate the risk of each vulnerability. Statistical methodology and underlying mathematical approach is described. Initially, parametric procedures are conducted and measured. There were, however, violations of some statistical assumptions observed. Therefore the need for non-parametric approaches was recognized. 6838 vulnerabilities recorded were considered in the analysis. According to the risk associated with all the vulnerabilities considered, it was found that there is a statistically significant difference among average risk levels for some operating systems, indicating that according to our method some operating systems have been more risk vulnerable than others given the assumptions and limitations. Relevant test results revealing a statistically significant difference in the Risk levels of different OSs are presented.

Keywords: cybersecurity, Markov chain, non-parametric analysis, vulnerability, operating system

Procedia PDF Downloads 159

Authors: Vanessa Cho, Janet Hsiao, Nigel King, Robert Anthonappa

Abstract:

Background: The personal protective equipment (PPE) requirements for health care workers (HCWs) have changed significantly during the COVID-19 pandemic. Aim: To ascertain, using eye-tracking technology, what children notice the most when seeing HCWs in various PPE. Design: A Tobii nano pro-eye-tracking camera tracked 156 children's visual attention while they viewed photographs of HCWs in various PPEs. Eye Movement analysis with Hidden Markov Models (EMHMM) was employed to analyse 624 recordings using two approaches, namely (i) data-driven where children's fixation determined the regions of interest (ROIs), and (ii) fixed ROIs where the investigators predefined the ROIs. Results: Two significant eye movement patterns, namely distributed(85.2%) and selective(14.7%), were identified(P<0.05). Most children fixated primarily on the face regardless of the different PPEs. Children fixated equally on all PPE images in the distributed pattern, while a strong preference for unmasked faces was evident in the selective pattern (P<0.01). Conclusion: Children as young as 2.5 years used a top-down visual search behaviour and demonstrated their face processing ability. Most children did not show a strong visual preference for a specific PPE, while a minority preferred PPE with distinct facial features, namely without masks and loupes.

Keywords: COVID-19, PPE, dentistry, pediatric

Procedia PDF Downloads 59

Authors: Omar Alzeley, Sergey Utev

Abstract:

Minimizing a constrained multivariate function is the fundamental of Machine learning, and these algorithms are at the core of data mining and data visualization techniques. The decision function that maps input points to output points is based on the result of optimization. This optimization is the central of learning theory. One approach to complex systems where the dynamics of the system is inferred by a statistical analysis of the fluctuations in time of some associated observable is time series analysis. The purpose of this paper is a mathematical transition from the autoregressive model of classical time series to the matrix formalization of quantum theory. Firstly, we have proposed a quantum time series model (QTS). Although Hamiltonian technique becomes an established tool to detect a deterministic chaos, other approaches emerge. The quantum probabilistic technique is used to motivate the construction of our QTS model. The QTS model resembles the quantum dynamic model which was applied to financial data. Secondly, various statistical methods, including machine learning algorithms such as the Kalman filter algorithm, are applied to estimate and analyses the unknown parameters of the model. Finally, simulation techniques such as Markov chain Monte Carlo have been used to support our investigations. The proposed model has been examined by using real and simulated data. We establish the relation between quantum statistical machine and quantum time series via random matrix theory. It is interesting to note that the primary focus of the application of QTS in the field of quantum chaos was to find a model that explain chaotic behaviour. Maybe this model will reveal another insight into quantum chaos.

Keywords: machine learning, simulation techniques, quantum probability, tensor product, time series

Procedia PDF Downloads 437

Authors: Ayesha Selwyn, Annapoorni Chandrashekar, Kumar Ashwarya, Nishant Baghel

Abstract:

Project-based learning (PBL) activities are typically implemented in technology-enabled schools by highly trained teachers. In rural India, students have limited access to technology and quality education. Implementing typical PBL activities is challenging. This study details how Pratham Education Foundation’s Hybrid Learning model was used to implement two PBL activities related to music in 900 remote Indian villages with 46,000 students aged 10-14. The activities were completed by 69% of groups that submitted a total of 15,000 videos (completed projects). Pratham’s H-Learning model reaches 100,000 students aged 3-14 in 900 Indian villages. The community-driven model engages students in 20,000 self-organized groups outside of school. The students are guided by 6,000 youth volunteers and 100 facilitators. The students partake in learning activities across subjects with the support of community stakeholders and offline digital content on shared Android tablets. A training and implementation toolkit for PBL activities is designed by subject experts. This toolkit is essential in ensuring efficient implementation of activities as facilitators aren’t highly skilled and have limited access to training resources. The toolkit details the activity at three levels of student engagement - enrollment, participation, and completion. The subject experts train project leaders and facilitators who train youth volunteers. Volunteers need to be trained on how to execute the activity and guide students. The training is focused on building the volunteers’ capacity to enable students to solve problems, rather than developing the volunteers’ subject-related knowledge. This structure ensures that continuous intervention of subject matter experts isn’t required, and the onus of judging creativity skills is put on community members. 46,000 students in the H-Learning program were engaged in two PBL activities related to Music from April-June 2019. For one activity, students had to conduct a “musical survey” in their village by designing a survey and shooting and editing a video. This activity aimed to develop students’ information retrieval, data gathering, teamwork, communication, project management, and creativity skills. It also aimed to identify talent and document local folk music. The second activity, “Pratham Idol”, was a singing competition. Students participated in performing, producing, and editing videos. This activity aimed to develop students’ teamwork and creative skills and give students a creative outlet. Students showcased their completed projects at village fairs wherein a panel of community members evaluated the videos. The shortlisted videos from all villages were further evaluated by experts who identified students and adults to participate in advanced music workshops. The H-Learning framework enables students in low resource settings to engage in PBL and develop relevant skills by leveraging community support and using video creation as a tool. In rural India, students do not have access to high-quality education or infrastructure. Therefore designing activities that can be implemented by community members after limited training is essential. The subject experts have minimal intervention once the activity is initiated, which significantly reduces the cost of implementation and allows the activity to be implemented at a massive scale.

Keywords: community supported learning, project-based learning, self-organized learning, education technology

Procedia PDF Downloads 152

Authors: Alinka Molnár-Tóth, Tímea Tánczos, Regina Barna, Katalin Jakab, Péter Klivényi

Abstract:

Supported by neuroscientific findings, the so-called Hub-and-Spoke model of the human semantic system is based on two subcomponents of semantic cognition, namely the semantic control process and semantic representation. Our semantic knowledge is multimodal in nature, as the knowledge system stored in relation to a conception is extensive and broad, while different aspects of the conception may be relevant depending on the purpose. The motivation of our research is to develop a new diagnostic measurement procedure based on the preservation of semantic representation, which is appropriate to the specificities of the Hungarian language and which can be used to compare the non-verbal semantic knowledge of healthy and aphasic persons. The development of the test will broaden the Hungarian clinical diagnostic toolkit, which will allow for more specific therapy planning. The sample of healthy persons (n=480) was determined by the last census data for the representativeness of the sample. Based on the concept of the Pyramids and Palm Tree Test, and according to the characteristics of the Hungarian language, we have elaborated a test based on different types of semantic information, in which the subjects are presented with three pictures: they have to choose the one that best fits the target word above from the two lower options, based on the semantic relation defined. We have measured 5 types of semantic knowledge representations: associative relations, taxonomy, motional representations, concrete as well as abstract verbs. As the first step in our data analysis, we examined the normal distribution of our results, and since it was not normally distributed (p < 0.05), we used nonparametric statistics further into the analysis. Using descriptive statistics, we could determine the frequency of the correct and incorrect responses, and with this knowledge, we could later adjust and remove the items of questionable reliability. The reliability was tested using Cronbach’s α, and it can be safely said that all the results were in an acceptable range of reliability (α = 0.6-0.8). We then tested for the potential gender differences using the Mann Whitney-U test, however, we found no difference between the two (p < 0.05). Likewise, we didn’t see that the age had any effect on the results using one-way ANOVA (p < 0.05), however, the level of education did influence the results (p > 0.05). The relationships between the subtests were observed by the nonparametric Spearman’s rho correlation matrix, showing statistically significant correlation between the subtests (p > 0.05), signifying a linear relationship between the measured semantic functions. A margin of error of 5% was used in all cases. The research will contribute to the expansion of the clinical diagnostic toolkit and will be relevant for the individualised therapeutic design of treatment procedures. The use of a non-verbal test procedure will allow an early assessment of the most severe language conditions, which is a priority in the differential diagnosis. The measurement of reaction time is expected to advance prodrome research, as the tests can be easily conducted in the subclinical phase.

Keywords: communication disorders, diagnostic toolkit, neurorehabilitation, semantic knowlegde

Procedia PDF Downloads 74

Authors: Jefferson Hernandez, Juan Padilla

Abstract:

Estimation of price elasticity of demand is a valuable tool for the task of price settling. Given its relevance, it is an active field for microeconomic and statistical research. Price elasticity in the industry of oil and gas, in particular for fuels sold in gas stations, has shown to be a challenging topic given the market and state restrictions, and underlying correlations structures between the types of fuels sold by the same gas station. This paper explores the Lotka-Volterra model for the problem for price elasticity estimation in the context of fuels; in addition, it is introduced multivariate random effects with the purpose of dealing with errors, e.g., measurement or missing data errors. In order to model the underlying correlation structures, the Inverse-Wishart, Hierarchical Half-t and LKJ distributions are studied. Here, the Bayesian paradigm through Markov Chain Monte Carlo (MCMC) algorithms for model estimation is considered. Simulation studies covering a wide range of situations were performed in order to evaluate parameter recovery for the proposed models and algorithms. Results revealed that the proposed algorithms recovered quite well all model parameters. Also, a real data set analysis was performed in order to illustrate the proposed approach.

Keywords: price elasticity, volume, correlation structures, Bayesian models

Procedia PDF Downloads 130

Authors: Mehrnoosh Omati, Mahmod Reza Sahebi

Abstract:

The information on land use/land cover changing plays an essential role for environmental assessment, planning and management in regional development. Remotely sensed imagery is widely used for providing information in many change detection applications. Polarimetric Synthetic aperture radar (PolSAR) image, with the discrimination capability between different scattering mechanisms, is a powerful tool for environmental monitoring applications. This paper proposes a new boundary-based segmentation algorithm as a fundamental step for land cover change detection. In this method, first, two PolSAR images are segmented using integration of marker-controlled watershed algorithm and coupled Markov random field (MRF). Then, object-based classification is performed to determine changed/no changed image objects. Compared with pixel-based support vector machine (SVM) classifier, this novel segmentation algorithm significantly reduces the speckle effect in PolSAR images and improves the accuracy of binary classification in object-based level. The experimental results on Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) polarimetric images show a 3% and 6% improvement in overall accuracy and kappa coefficient, respectively. Also, the proposed method can correctly distinguish homogeneous image parcels.

Keywords: coupled Markov random field (MRF), environment, object-based analysis, polarimetric SAR (PolSAR) images

Procedia PDF Downloads 196

Authors: Xinyuan Y. Song, Kai Kang

Abstract:

Joint models for analyzing longitudinal and survival data are widely used to investigate the relationship between a failure time process and time-variant predictors. A common assumption in conventional joint models in the survival analysis literature is that all predictors are observable. However, this assumption may not always be supported because unobservable traits, namely, latent variables, which are indirectly observable and should be measured through multiple observed variables, are commonly encountered in the medical, behavioral, and financial research settings. In this study, a joint modeling approach to deal with this feature is proposed. The proposed model comprises three parts. The first part is a dynamic factor analysis model for characterizing latent variables through multiple observed indicators over time. The second part is a random coefficient trajectory model for describing the individual trajectories of latent variables. The third part is a proportional hazard model for examining the effects of time-invariant predictors and the longitudinal trajectories of time-variant latent risk factors on hazards of interest. A Bayesian approach coupled with a Markov chain Monte Carlo algorithm to perform statistical inference. An application of the proposed joint model to a study on the Alzheimer's disease neuroimaging Initiative is presented.

Keywords: Bayesian analysis, joint model, longitudinal data, time-to-event data

Procedia PDF Downloads 116

Authors: Ilker Ali Ozkan

Abstract:

In this study, an artificial neural network model has been developed to estimate chronic kidney failure which is a common disease. The patients’ age, their blood and biochemical values, and 24 input data which consists of various chronic diseases are used for the estimation process. The input data have been subjected to preprocessing because they contain both missing values and nominal values. 147 patient data which was obtained from the preprocessing have been divided into as 70% training and 30% testing data. As a result of the study, artificial neural network model with 25 neurons in the hidden layer has been found as the model with the lowest error value. Chronic kidney failure disease has been able to be estimated accurately at the rate of 99.3% using this artificial neural network model. The developed artificial neural network has been found successful for the estimation of chronic kidney failure disease using clinical data.

Keywords: estimation, artificial neural network, chronic kidney failure disease, disease diagnosis

Procedia PDF Downloads 416

Authors: Tet H. Yeap

Abstract:

An analog restricted Hopfield Network is presented in this paper. It consists of two layers of nodes, visible and hidden nodes, connected by directional weighted paths forming a bipartite graph with no intralayer connection. An energy or Lyapunov function was derived to show that the proposed network will converge to stable states. By introducing hidden nodes, the proposed network can be trained to store patterns and has increased memory capacity. Training to be an associative memory, simulation results show that the associative memory performs better than a classical Hopfield network by being able to perform better memory recall when the input is noisy.

Keywords: restricted Hopfield network, Lyapunov function, simultaneous perturbation stochastic approximation

Procedia PDF Downloads 105

Authors: Ramanpreet Kaur, Upasana Sharma

Abstract:

The purpose of this paper is to address the challenges facing the water supply for the Machine Learning Database (MLDB) system at the piston foundry plant. In the MLDB system, one main unit, i.e., robotic, is connected by two sub-units. The functioning of the system depends on the robotic and water supply. Lack of water supply causes system failure. The system operates at full capacity with the help of two sub-units. If one sub-unit fails, the system runs at a low capacity. Reliability modeling is performed using semi-Markov processes and regenerative point techniques. Several system effects such as mean time to system failure, availability at full capacity, availability at reduced capacity, busy period for repair and expected number of visits have been achieved. Benefits have been analyzed. The graphical study is designed for a specific case using programming in C++ and MS Excel.

Keywords: MLDB system, robotic, semi-Markov process, regenerative point technique

Procedia PDF Downloads 82

Authors: Suparman

Abstract:

Piecewise linear regression models are very flexible models for modeling the data. If the piecewise linear regression models are matched against the data, then the parameters are generally not known. This paper studies the problem of parameter estimation of piecewise linear regression models. The method used to estimate the parameters of picewise linear regression models is Bayesian method. But the Bayes estimator can not be found analytically. To overcome these problems, the reversible jump MCMC algorithm is proposed. Reversible jump MCMC algorithm generates the Markov chain converges to the limit distribution of the posterior distribution of the parameters of picewise linear regression models. The resulting Markov chain is used to calculate the Bayes estimator for the parameters of picewise linear regression models.

Keywords: regression, piecewise, Bayesian, reversible Jump MCMC

Procedia PDF Downloads 492

Authors: Amr Sadek, Abdelrahaman Al-Qahtany, Turkey Salem Al-Qahtany

Abstract:

In this study, a backpropagation artificial neural network (ANN) model has been used to predict the cooling and heating capacities of air conditioners (AC) under different conditions. Sufficiently large measurement results were obtained from the national energy-efficiency laboratories in Saudi Arabia and were used for the learning process of the ANN model. The parameters affecting the performance of the AC, including temperature, humidity level, specific heat enthalpy indoors and outdoors, and the air volume flow rate of indoor units, have been considered. These parameters were used as inputs for the ANN model, while the cooling and heating capacity values were set as the targets. A backpropagation ANN model with two hidden layers and one output layer could successfully correlate the input parameters with the targets. The characteristics of the ANN model including the input-processing, transfer, neurons-distance, topology, and training functions have been discussed. The performance of the ANN model was monitored over the training epochs and assessed using the mean squared error function. The model was then used to predict the performance of the AC under conditions that were not included in the measurement results. The optimum performance of the AC was also predicted under the different environmental conditions in Saudi Arabia. The uncertainty of the ANN model predictions has been evaluated taking into account the randomness of the data and lack of learning.

Keywords: artificial neural network, uncertainty of model predictions, efficiency of air conditioners, cooling and heating capacities

Procedia PDF Downloads 44

Authors: M. Kastek, M. Kowalski, M. Szustakowski, H. Polakowski, T. Sosnowski

Abstract:

Development of highly efficient security systems is one of the most urgent topics for science and engineering. There are many kinds of threats and many methods of prevention. It is very important to detect a threat as early as possible in order to neutralize it. One of the very challenging problems is detection of dangerous objects hidden under human’s clothing. This problem is particularly important for safety of airport passengers. In order to develop methods and algorithms to detect hidden objects it is necessary to determine the thermal signatures of such objects of interest. The laboratory measurements were conducted to determine the thermal signatures of dangerous tools hidden under various clothes in different ambient conditions. Cameras used for measurements were working in spectral range 0.6-12.5 μm An infrared imaging Fourier transform spectroradiometer was also used, working in spectral range 7.7-11.7 μm. Analysis of registered thermograms and hyperspectral datacubes has yielded the thermal signatures for two types of guns, two types of knives and home-made explosive bombs. The determined thermal signatures will be used in the development of method and algorithms of image analysis implemented in proposed monitoring systems.

Keywords: hyperspectral detection, nultispectral detection, image processing, monitoring systems

Procedia PDF Downloads 323

Authors: Min Jindong, Wang Mingxia

Abstract:

With the widespread use of artificial intelligence in life, computer vision, especially deep convolutional neural network models, has developed rapidly. With the increase of the complexity of the real visual target detection task and the improvement of the recognition accuracy, the target detection network model is also very large. The huge deep neural network model is not conducive to deployment on edge devices with limited resources, and the timeliness of network model inference is poor. In this paper, knowledge distillation is used to compress the huge and complex deep neural network model, and the knowledge contained in the complex network model is comprehensively transferred to another lightweight network model. Different from traditional knowledge distillation methods, we propose a novel knowledge distillation that incorporates multi-faceted features, called M-KD. In this paper, when training and optimizing the deep neural network model for target detection, the knowledge of the soft target output of the teacher network in knowledge distillation, the relationship between the layers of the teacher network and the feature attention map of the hidden layer of the teacher network are transferred to the student network as all knowledge. in the model. At the same time, we also introduce an intermediate transition layer, that is, an intermediate guidance layer, between the teacher network and the student network to make up for the huge difference between the teacher network and the student network. Finally, this paper adds an exploration module to the traditional knowledge distillation teacher-student network model. The student network model not only inherits the knowledge of the teacher network but also explores some new knowledge and characteristics. Comprehensive experiments in this paper using different distillation parameter configurations across multiple datasets and convolutional neural network models demonstrate that our proposed new network model achieves substantial improvements in speed and accuracy performance.

Keywords: object detection, knowledge distillation, convolutional network, model compression

Procedia PDF Downloads 251

Authors: Ágoston Nagy

Abstract:

This paper introduces a method of using unsupervised machine learning to visualize the feature space of a dataset in 2D, in order to find most characteristic segments in the set. After dimension reduction, users can select clusters by manual drawing. Selected clusters are recorded into a data model that is used for later predictions, based on realtime data. Predictions are made with supervised learning, using Gesture Recognition Toolkit. The paper introduces two example applications: a semantic audio organizer for analyzing incoming sounds, and a gesture database organizer where gestural data (recorded by a Leap motion) is visualized for further manipulation.

Keywords: gesture recognition, machine learning, real-time interaction, visualization

Procedia PDF Downloads 324

Authors: P. C. Tewari, Parveen Kumar

Abstract:

The paper describes the availability analysis of milling system of a rice milling plant using probabilistic approach. The subsystems under study are special purpose machines. The availability analysis of the system is carried out to determine the effect of failure and repair rates of each subsystem on overall performance (i.e. steady state availability) of system concerned. Further, on the basis of effect of repair rates on the system availability, maintenance repair priorities have been suggested. The problem is formulated using Markov Birth-Death process taking exponential distribution for probable failures and repair rates. The first order differential equations associated with transition diagram are developed by using mnemonic rule. These equations are solved using normalizing conditions and recursive method to drive out the steady state availability expression of the system. The findings of the paper are presented and discussed with the plant personnel to adopt a suitable maintenance policy to increase the productivity of the rice milling plant.

Keywords: availability modeling, Markov process, milling system, rice milling plant

Procedia PDF Downloads 206

Authors: Kulwinder Singh Mann, Manmohan Singh Heer, Asha Rani

Abstract:

The safe storage of radioactive materials has become an important issue. Nuclear engineering necessitates the safe handling of radioactive materials emitting high energy gamma-rays. Hazards involved in handling radioactive materials insist suitable shielded enclosures. With overgrowing use of nuclear energy for meeting the increasing demand of power, there is a need to investigate the shielding behavior of cost effective shielded enclosure (CESE) made from clay-bricks (CB) and fire-bricks (FB). In comparison to the lead-bricks (conventional-shielding), the CESE are the preferred choice in nuclear waste management. The objective behind the present investigation is to evaluate the double layered transmission exposure buildup factors (DLEBF) for gamma-rays for CESE in energy range 0.5-3MeV. For necessary computations of shielding parameters, using existing huge data regarding gamma-rays interaction parameters of all periodic table elements, two computer programs (GRIC-toolkit and BUF-toolkit) have been designed. It has been found that two-layered slabs show effective shielding for gamma-rays in orientation CB followed by FB than the reverse. It has been concluded that the arrangement, FB followed by CB reduces the leakage of scattered gamma-rays from the radioactive source.

Keywords: buildup factor, clay bricks, fire bricks, nuclear wastage management, radiation protective double layered slabs

Procedia PDF Downloads 379

Authors: M. Draman, S. Z. Muhamad Yassin, M. S. Alias, Z. Lambak, M. I. Zulkifli, S. N. Padhi, K. N. Baharim, F. Maskuriy, A. I. A. Rahim

Abstract:

The performance of Malay automatic speech recognition (ASR) system for the call centre environment is presented. The system utilizes Kaldi toolkit as the platform to the entire library and algorithm used in performing the ASR task. The acoustic model implemented in this system uses a deep neural network (DNN) method to model the acoustic signal and the standard (n-gram) model for language modelling. With 80 hours of training data from the call centre recordings, the ASR system can achieve 72% of accuracy that corresponds to 28% of word error rate (WER). The testing was done using 20 hours of audio data. Despite the implementation of DNN, the system shows a low accuracy owing to the varieties of noises, accent and dialect that typically occurs in Malaysian call centre environment. This significant variation of speakers is reflected by the large standard deviation of the average word error rate (WERav) (i.e., ~ 10%). It is observed that the lowest WER (13.8%) was obtained from recording sample with a standard Malay dialect (central Malaysia) of native speaker as compared to 49% of the sample with the highest WER that contains conversation of the speaker that uses non-standard Malay dialect.

Keywords: conversational speech recognition, deep neural network, Malay language, speech recognition

Procedia PDF Downloads 297

Authors: Wanhyun Cho, Soonja Kang, Sangkyoon Kim, Soonyoung Park

Abstract:

In this paper, we present the human action recognition method using the variational Bayesian HMM with the Dirichlet process mixture (DPM) of the Gaussian-Wishart emission model (GWEM). First, we define the Bayesian HMM based on the Dirichlet process, which allows an infinite number of Gaussian-Wishart components to support continuous emission observations. Second, we have considered an efficient variational Bayesian inference method that can be applied to drive the posterior distribution of hidden variables and model parameters for the proposed model based on training data. And then we have derived the predictive distribution that may be used to classify new action. Third, the paper proposes a process of extracting appropriate spatial-temporal feature vectors that can be used to recognize a wide range of human behaviors from input video image. Finally, we have conducted experiments that can evaluate the performance of the proposed method. The experimental results show that the method presented is more efficient with human action recognition than existing methods.

Keywords: human action recognition, Bayesian HMM, Dirichlet process mixture model, Gaussian-Wishart emission model, Variational Bayesian inference, prior distribution and approximate posterior distribution, KTH dataset

Procedia PDF Downloads 323

Authors: Leonard Maaya, Michel Meulders, Martina Vandebroek

Abstract:

Discrete Choice Experiments (DCE) are used to investigate how product attributes affect decision-makers’ choices. In DCEs, choice situations consisting of several alternatives are presented from which choice-makers select the preferred alternative. Standard multinomial logit models based on random utility theory can be used to estimate the utilities for the attributes. The overarching principle in these models is that respondents understand and use all the attributes when making choices. However, studies suggest that respondents sometimes ignore some attributes (commonly referred to as Attribute Non-Attendance/ANA). The choice modeling literature presents ANA as a static process, i.e., respondents’ ANA behavior does not change throughout the experiment. However, respondents may ignore attributes due to changing factors like availability of information on attributes, learning/fatigue in experiments, etc. We develop a dynamic mixture latent Markov model to model changes in ANA when information on attributes is provided. The model is illustrated on e-consumers’ webshop choices. The results indicate that the dynamic ANA model describes the behavioral changes better than modeling the impact of information using changes in parameters. Further, we find that providing information on attributes leads to an increase in the attendance probabilities for the investigated attributes.

Keywords: choice models, discrete choice experiments, dynamic models, e-commerce, statistical modeling

Procedia PDF Downloads 110

Authors: E. A. Tkachenko, E. M. Rogova, V. V. Klimov

Abstract:

One of the basic issues of development management is connected with performance measurement as a prerequisite for identifying the achievement of development objectives. The aim of our research is to develop an improved model of assessing a company’s development results. The model should take into account the cyclical nature of development and the high degree of uncertainty in dealing with numerous management tasks. Our hypotheses may be formulated as follows: Hypothesis 1. The cycle of a company’s development may be studied from the standpoint of a project cycle. To do that, methods and tools of project analysis are to be used. Hypothesis 2. The problem of the uncertainty when justifying managerial decisions within the framework of a company’s development cycle can be solved through the use of the mathematical apparatus of fuzzy logic. The reasoned justification of the validity of the hypotheses made is given in the suggested article. The fuzzy logic toolkit applies to the case of technology shift within an enterprise. It is proven that some restrictions in performance measurement that are incurred to conventional methods could be eliminated by implementation of the fuzzy logic apparatus in performance measurement models.

Keywords: logic, fuzzy sets, performance measurement, project analysis

Procedia PDF Downloads 349

Authors: Waleed Basuliman

Abstract:

Anthropometric dimensions are considered one of the important factors when designing human-machine systems. In this study, the estimation of anthropometric dimensions has been improved by using Artificial Neural Network (ANN) model that is able to predict the anthropometric measurements of Saudi males in Riyadh City. A total of 1427 Saudi males aged 6 to 60 years participated in measuring 20 anthropometric dimensions. These anthropometric measurements are considered important for designing the work and life applications in Saudi Arabia. The data were collected during eight months from different locations in Riyadh City. Five of these dimensions were used as predictors variables (inputs) of the model, and the remaining 15 dimensions were set to be the measured variables (Model’s outcomes). The hidden layers varied during the structuring stage, and the best performance was achieved with the network structure 6-25-15. The results showed that the developed Neural Network model was able to estimate the body dimensions of Saudi male population in Riyadh City. The network's mean absolute percentage error (MAPE) and the root mean squared error (RMSE) were found to be 0.0348 and 3.225, respectively. These results were found less, and then better, than the errors found in the literature. Finally, the accuracy of the developed neural network was evaluated by comparing the predicted outcomes with regression model. The ANN model showed higher coefficient of determination (R2) between the predicted and actual dimensions than the regression model.

Keywords: artificial neural network, anthropometric measurements, back-propagation

Procedia PDF Downloads 464

Authors: Yina A. Cifuentes Triana, Andrés M. Pinzón Velásco, Marío E. Velásquez Lozano

Abstract:

In this work the sequencing and genome characterization of a natural isolate of Saccharomyces cerevisiae yeast (strain 202-3), identified with potential for the production of second generation ethanol from sugarcane bagasse hydrolysates is presented. This strain was selected because its capability to consume xylose during the fermentation of sugarcane bagasse hydrolysates, taking into account that many strains of S. cerevisiae are incapable of processing this sugar. This advantage and other prominent positive aspects during fermentation profiles evaluated in bagasse hydrolysates made the strain 202-3 a candidate strain to improve the production of second-generation ethanol, which was proposed as a first step to study the strain at the genomic level. The molecular characterization was carried out by genome sequencing with the Illumina HiSeq 2000 platform paired end; the assembly was performed with different programs, finally choosing the assembler ABYSS with kmer 89. Gene prediction was developed with the approach of hidden Markov models with Augustus. The genes identified were scored based on similarity with public databases of nucleotide and protein. Records were organized from ontological functions at different hierarchical levels, which identified central metabolic functions and roles of the S. cerevisiae strain 202-3, highlighting the presence of four possible new proteins, two of them probably associated with the positive consumption of xylose.

Keywords: cellulosic ethanol, Saccharomyces cerevisiae, genome sequencing, xylose consumption

Procedia PDF Downloads 297

Authors: V. Makis, L. Jafari

Abstract:

In this paper, we develop an optimal Bayesian chart to control the expected number of defects per inspection unit in production processes with long production runs. We formulate this control problem in the optimal stopping framework. The objective is to determine the optimal stopping rule minimizing the long-run expected average cost per unit time considering partial information obtained from the process sampling at regular epochs. We prove the optimality of the control limit policy, i.e., the process is stopped and the search for assignable causes is initiated when the posterior probability that the process is out of control exceeds a control limit. An algorithm in the semi-Markov decision process framework is developed to calculate the optimal control limit and the corresponding average cost. Numerical examples are presented to illustrate the developed optimal control chart and to compare it with the traditional u-chart.

Keywords: Bayesian u-chart, economic design, optimal stopping, semi-Markov decision process, statistical process control

Procedia PDF Downloads 542

Authors: Joselito Medina-Marin, Maria G. Serna-Diaz, Alejandro Tellez-Jurado, Juan C. Seck-Tuoh-Mora, Eva S. Hernandez-Gress, Norberto Hernandez-Romero, Iaina P. Medina-Serna

Abstract:

Trichoderma harzianum is a fungus that has been utilized as a low-cost fungicide for biological control of pests, and it is important to determine the optimal conditions to produce the highest amount of conidiospores of Trichoderma harzianum. In this work, the conidiospore production of Trichoderma harzianum is modeled and optimized by using Artificial Neural Networks (AANs). In order to gather data of this process, 30 experiments were carried out taking into account the number of hours of culture (10 distributed values from 48 to 136 hours) and the culture humidity (70, 75 and 80 percent), obtained as a response the number of conidiospores per gram of dry mass. The experimental results were used to develop an iterative algorithm to create 1,110 ANNs, with different configurations, starting from one to three hidden layers, and every hidden layer with a number of neurons from 1 to 10. Each ANN was trained with the Levenberg-Marquardt backpropagation algorithm, which is used to learn the relationship between input and output values. The ANN with the best performance was chosen in order to simulate the process and be able to maximize the conidiospores production. The obtained ANN with the highest performance has 2 inputs and 1 output, three hidden layers with 3, 10 and 10 neurons in each layer, respectively. The ANN performance shows an R2 value of 0.9900, and the Root Mean Squared Error is 1.2020. This ANN predicted that 644175467 conidiospores per gram of dry mass are the maximum amount obtained in 117 hours of culture and 77% of culture humidity. In summary, the ANN approach is suitable to represent the conidiospores production of Trichoderma harzianum because the R2 value denotes a good fitting of experimental results, and the obtained ANN model was used to find the parameters to produce the biggest amount of conidiospores per gram of dry mass.

Keywords: Trichoderma harzianum, modeling, optimization, artificial neural network

Procedia PDF Downloads 126

Authors: Catia S. A. Sima, K. Bousson

Abstract:

Congestion in the Terminal Maneuvering Areas (TMAs) of larger airports impacts all aspects of air traffic flow, not only at national level but may also induce arrival delays at international level. Hence, there is a need to monitor appropriately the air traffic flow in TMAs so that efficient decisions may be taken to manage their occupancy rates. It would be desirable to physically increase the existing airspace to accommodate all existing demands, but this question is entirely utopian and, given this possibility, several studies and analyses have been developed over the past decades to meet the challenges that have arisen due to the dizzying expansion of the aeronautical industry. The main objective of the present paper is to propose concepts to manage and reduce the degree of uncertainty in the air traffic operations, maximizing the interest of all involved, ensuring a balance between demand and supply, and developing and/or adapting resources that enable a rapid and effective adaptation of measures to the current context and the consequent changes perceived in the aeronautical industry. A central task is to emphasize the increase in air traffic flow management capacity to the present day, taking into account not only a wide range of methodologies but also equipment and/or tools already available in the aeronautical industry. The efficient use of these resources is crucial as the human capacity for work is limited and the actors involved in all processes related to air traffic flow management are increasingly overloaded and, as a result, operational safety could be compromised. The methodology used to answer and/or develop the issues listed above is based on the advantages promoted by the application of Markov Chain principles that enable the construction of a simplified model of a dynamic network that describes the air traffic flow behavior anticipating their changes and eventual measures that could better address the impact of increased demand. Through this model, the proposed concepts are shown to have potentials to optimize the air traffic flow management combined with the operation of the existing resources at each moment and the circumstances found in each TMA, using historical data from the air traffic operations and specificities found in the aeronautical industry, namely in the Portuguese context.

Keywords: air traffic flow, terminal maneuvering area, TMA, air traffic management, ATM, Markov chains

Procedia PDF Downloads 106