Search results for: latent variable model

Authors: Jiyoung Bae, Gyoomi Kim

Abstract:

This study explores a range of linguistic features used in Korean college students’ argumentative writings for the purpose of developing a model that identifies variables which predict writing proficiencies. This study investigated the latent variable structure of L2 linguistic features, including syntactic complexity, the lexical complexity, and fluency. One hundred forty-six university students in Korea participated in this study. The results of the study’s confirmatory factor analysis (CFA) showed that indicators of linguistic features from this study-provided a foundation for re-categorizing indicators found in extant research on L2 Korean writers depending on each latent variable of linguistic features. The CFA models indicated one measurement model of L2 syntactic complexity and L2 learners’ writing proficiency; these two latent factors were correlated with each other. Based on the overall findings of the study, integrated linguistic features of L2 writings suggested some pedagogical implications in L2 writing instructions.

Keywords: linguistic features, syntactic complexity, lexical complexity, fluency

Procedia PDF Downloads 148

Authors: E. Tchandao Mangamana, V. Cariou, E. Vigneau, R. Glele Kakai, E. M. Qannari

Abstract:

A new definition of a latent variable associated with a dataset makes it possible to propose variants of the PLS2 regression and the multi-block PLS (MB-PLS). We shall refer to these variants as Rd-PLS regression and Rd-MB-PLS respectively because they are inspired by both Redundancy analysis and PLS regression. Usually, a latent variable t associated with a dataset Z is defined as a linear combination of the variables of Z with the constraint that the length of the loading weights vector equals 1. Formally, t=Zw with ‖w‖=1. Denoting by Z' the transpose of Z, we define herein, a latent variable by t=ZZ’q with the constraint that the auxiliary variable q has a norm equal to 1. This new definition of a latent variable entails that, as previously, t is a linear combination of the variables in Z and, in addition, the loading vector w=Z’q is constrained to be a linear combination of the rows of Z. More importantly, t could be interpreted as a kind of projection of the auxiliary variable q onto the space generated by the variables in Z, since it is collinear to the first PLS1 component of q onto Z. Consider the situation in which we aim to predict a dataset Y from another dataset X. These two datasets relate to the same individuals and are assumed to be centered. Let us consider a latent variable u=YY’q to which we associate the variable t= XX’YY’q. Rd-PLS consists in seeking q (and therefore u and t) so that the covariance between t and u is maximum. The solution to this problem is straightforward and consists in setting q to the eigenvector of YY’XX’YY’ associated with the largest eigenvalue. For the determination of higher order components, we deflate X and Y with respect to the latent variable t. Extending Rd-PLS to the context of multi-block data is relatively easy. Starting from a latent variable u=YY’q, we consider its ‘projection’ on the space generated by the variables of each block Xk (k=1, ..., K) namely, tk= XkXk'YY’q. Thereafter, Rd-MB-PLS seeks q in order to maximize the average of the covariances of u with tk (k=1, ..., K). The solution to this problem is given by q, eigenvector of YY’XX’YY’, where X is the dataset obtained by horizontally merging datasets Xk (k=1, ..., K). For the determination of latent variables of order higher than 1, we use a deflation of Y and Xk with respect to the variable t= XX’YY’q. In the same vein, extending Rd-MB-PLS to the path modeling setting is straightforward. Methods are illustrated on the basis of case studies and performance of Rd-PLS and Rd-MB-PLS in terms of prediction is compared to that of PLS2 and MB-PLS.

Keywords: multiblock data analysis, partial least squares regression, path modeling, redundancy analysis

Procedia PDF Downloads 124

Authors: Zhen Cheng, Xinyu Dai, Shujian Huang, Jiajun Chen

Abstract:

Recently, explanatory natural language inference has attracted much attention for the interpretability of logic relationship prediction, which is also known as explanation generation for Natural Language Inference (NLI). Existing explanation generators based on discriminative Encoder-Decoder architecture have achieved noticeable results. However, we find that these discriminative generators usually generate explanations with correct evidence but incorrect logic semantic. It is due to that logic information is implicitly encoded in the premise-hypothesis pairs and difficult to model. Actually, logic information identically exists between premise-hypothesis pair and explanation. And it is easy to extract logic information that is explicitly contained in the target explanation. Hence we assume that there exists a latent space of logic information while generating explanations. Specifically, we propose a generative model called Variational Explanation Generator (VariationalEG) with a latent variable to model this space. Training with the guide of explicit logic information in target explanations, latent variable in VariationalEG could capture the implicit logic information in premise-hypothesis pairs effectively. Additionally, to tackle the problem of posterior collapse while training VariaztionalEG, we propose a simple yet effective approach called Logic Supervision on the latent variable to force it to encode logic information. Experiments on explanation generation benchmark—explanation-Stanford Natural Language Inference (e-SNLI) demonstrate that the proposed VariationalEG achieves significant improvement compared to previous studies and yields a state-of-the-art result. Furthermore, we perform the analysis of generated explanations to demonstrate the effect of the latent variable.

Keywords: natural language inference, explanation generation, variational auto-encoder, generative model

Procedia PDF Downloads 128

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 126

Authors: Malika Yaici, Kamel Hariche, Tim Clarke

Abstract:

The relationship between eigenstructure (eigenvalues and eigenvectors) and latent structure (latent roots and latent vectors) is established. In control theory eigenstructure is associated with the state space description of a dynamic multi-variable system and a latent structure is associated with its matrix fraction description. Beginning with block controller and block observer state space forms and moving on to any general state space form, we develop the identities that relate eigenvectors and latent vectors in either direction. Numerical examples illustrate this result. A brief discussion of the potential of these identities in linear control system design follows. Additionally, we present a consequent result: a quick and easy method to solve the polynomial eigenvalue problem for regular matrix polynomials.

Keywords: eigenvalues/eigenvectors, latent values/vectors, matrix fraction description, state space description

Procedia PDF Downloads 453

Authors: Levent Dumenci, Laura A. Siminoff

Abstract:

Health literacy has been shown to predict a variety of health outcomes. Reliable identification of persons with limited cancer health literacy (LCHL) has been proved questionable with existing instruments using an arbitrary cut point along a continuum. The CHLT-6, however, uses a latent mixture modeling approach to identify persons with LCHL. The purpose of this study was to estimate two-week and six-month stability of identifying persons with LCHL using the CHLT-6 with a discrete latent variable approach as the underlying measurement structure. Using a test-retest design, the CHLT-6 was administered to cancer patients with two-week (N=98) and six-month (N=51) intervals. The two-week and six-month latent test-retest agreements were 89% and 88%, respectively. The chance-corrected latent agreements estimated from Dumenci’s latent kappa were 0.62 (95% CI: 0.41 – 0.82) and .47 (95% CI: 0.14 – 0.80) for the two-week and six-month intervals, respectively. High levels of latent test-retest agreement between limited and adequate categories of cancer health literacy construct, coupled with moderate to good levels of change-corrected latent agreements indicated that the CHLT-6 classification of limited versus adequate cancer health literacy is relatively stable over time. In conclusion, the measurement structure underlying the instrument allows for estimating classification errors circumventing limitations due to arbitrary approaches adopted by all other instruments. The CHLT-6 can be used to identify persons with LCHL in oncology clinics and intervention studies to accurately estimate treatment effectiveness.

Keywords: limited cancer health literacy, the CHLT-6, discrete latent variable modeling, latent agreement

Procedia PDF Downloads 162

Authors: Catalina Sau Man Ng

Abstract:

Workplace bullying is generally defined as a form of direct and indirect maltreatment at work including harassing, offending, socially isolating someone or negatively affecting someone’s work tasks. Workplace bullying is unfortunately commonplace around the world, which makes it a social phenomenon worth researching. However, the measurements and estimation methods of workplace bullying seem to be diverse in different studies, leading to dubious results. Hence, this paper attempts to examine the prevalence of workplace bullying in Hong Kong using the latent class analysis approach. It is often argued that the traditional classification of workplace bullying into the dichotomous 'victims' and 'non-victims' may not be able to fully represent the complex phenomenon of bullying. By treating workplace bullying as one latent variable and examining the potential categorical distribution within the latent variable, a more thorough understanding of workplace bullying in real-life situations may hence be provided. As a result, this study adopts a latent class analysis method, which was tested to demonstrate higher construct and higher predictive validity previously. In the present study, a representative sample of 2814 employees (Male: 54.7%, Female: 45.3%) in Hong Kong was recruited. The participants were asked to fill in a self-reported questionnaire which included measurements such as Chinese Workplace Bullying Scale (CWBS) and Chinese Version of Depression Anxiety Stress Scale (DASS). It is estimated that four latent classes will emerge: 'non-victims', 'seldom bullied', 'sometimes bullied', and 'victims'. The results of each latent class and implications of the study will also be discussed in this working paper.

Keywords: latent class analysis, prevalence, survey, workplace bullying

Procedia PDF Downloads 307

Authors: Phumelele Kubheka, Pius Owolawi, Gbolahan Aiyetoro

Abstract:

Twitter is one of the most popular social media platforms where users can share their opinions on different subjects. As of 2010, The Twitter platform generates more than 12 Terabytes of data daily, ~ 4.3 petabytes in a single year. For this reason, Twitter is a great source for big mining data. Many industries such as Telecommunication companies can leverage the availability of Twitter data to better understand their markets and make an appropriate business decision. This study performs topic modeling on Twitter data using Latent Dirichlet Allocation (LDA). The obtained results are benchmarked with another topic modeling technique, Latent Semantic Indexing (LSI). The study aims to retrieve topics on a Twitter dataset containing user tweets on South African Telcos. Results from this study show that LSI is much faster than LDA. However, LDA yields better results with higher topic coherence by 8% for the best-performing model represented in Table 1. A higher topic coherence score indicates better performance of the model.

Keywords: big data, latent Dirichlet allocation, latent semantic indexing, telco, topic modeling, twitter

Procedia PDF Downloads 136

Authors: Aswathi Thrivikraman, S. Advaith

Abstract:

The study is aimed at designing encoders for deciphering intricacies in time series data by redescribing the dynamics operating on a lower-dimensional manifold. A semi-supervised LSTM autoencoder is devised and investigated to see if the latent representation of the time series data can better forecast the data. End-to-end training of the LSTM autoencoder, together with another LSTM network that is connected to the latent space, forces the hidden states of the encoder to represent the most meaningful latent variables relevant for forecasting. Furthermore, the study compares the predictions with those of a traditional seq2seq model.

Keywords: LSTM, autoencoder, forecasting, seq2seq model

Procedia PDF Downloads 134

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

Abstract:

We present probabilistic multinomial Dirichlet classification model for multidimensional data and Gaussian process priors. Here, we have considered an efficient computational method that can be used to obtain the approximate posteriors for latent variables and parameters needed to define the multiclass Gaussian process classification model. We first investigated the process of inducing a posterior distribution for various parameters and latent function by using the variational Bayesian approximations and important sampling method, and next we derived a predictive distribution of latent function needed to classify new samples. The proposed model is applied to classify the synthetic multivariate dataset in order to verify the performance of our model. Experiment result shows that our model is more accurate than the other approximation methods.

Keywords: multinomial dirichlet classification model, Gaussian process priors, variational Bayesian approximation, importance sampling, approximate posterior distribution, marginal likelihood evidence

Procedia PDF Downloads 422

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 47

Authors: Shin-Hyung Cho, Dong-Kyu Kim, Seung-Young Kho

Abstract:

Truck-involved crashes have higher crash severity than non-truck-involved crashes. There have been many studies about the frequency of crashes and the development of severity models, but those studies only analyzed the relationship between observed variables. To identify why more people are injured or killed when trucks are involved in the crash, we must examine to quantify the complex causal relationship between severity of the crash and risk factors by adopting the latent factors of crashes. The aim of this study was to develop a structural equation or model based on truck-involved and non-truck-involved crashes, including five latent variables, i.e. a crash factor, environmental factor, road factor, driver’s factor, and severity factor. To clarify the unique characteristics of truck-involved crashes compared to non-truck-involved crashes, a confirmatory analysis method was used. To develop the model, we extracted crash data from 10,083 crashes on Korean freeways from 2008 through 2014. The results showed that the most significant variable affecting the severity of a crash is the crash factor, which can be expressed by the location, cause, and type of the crash. For non-truck-involved crashes, the crash and environment factors increase severity of the crash; conversely, the road and driver factors tend to reduce severity of the crash. For truck-involved crashes, the driver factor has a significant effect on severity of the crash although its effect is slightly less than the crash factor. The multiple group analysis employed to analyze the differences between the heterogeneous groups of drivers.

Keywords: crash severity, structural structural equation modeling (SEM), truck-involved crashes, multiple group analysis, crash on freeway

Procedia PDF Downloads 366

Authors: Pallavi Thakur, Rakesh K. Garg

Abstract:

Latent fingerprints are commonly encountered evidence at the scene of the crime. It is very important to decipher these fingerprints in order to explore their identity and a lot of research has been made on the visualization of latent fingermarks on various substrates by numerous researchers. During the past few years large number of powder formulations has been evolved for the development of latent fingermarks on different surfaces. This paper reports a new and simple fingerprint powder which is non-toxic and has been employed on different substrates successfully for the development and visualization of latent fingermarks upto the time period of twelve days in varying temperature conditions. In this study, a less expensive, simple and easily available catechu (kattha) powder has been used to decipher the latent fingermarks on different substrates namely glass, plastic, metal, aluminium foil, white paper, wall tile and wooden sheet. It is observed that it gives very clear results on all the mentioned substrates and can be successfully used for the development and visualization of twelve days old latent fingermarks in varying temperature conditions on wall tiles.

Keywords: fingermarks, catechu, visualization, aged fingermarks

Procedia PDF Downloads 175

Authors: Marco Sewald

Abstract:

Comparing present to past the numbers of Americans expatriating U. S. citizenship have risen. Even though these numbers are small compared to the immigrants, U. S. citizens expatriations have historically been much lower, making the uptick worrisome. In addition, the published lists and numbers from the U.S. government seems incomplete, with many not counted. Different branches of the U. S. government report different numbers and no one seems to know exactly how big the real number is, even though the IRS and the FBI both track and/or publish numbers of Americans who renounce. Since there is no single explanation, anecdotal evidence suggests this uptick is caused by global tax law and increased compliance burdens imposed by the U.S. lawmakers on U.S. citizens abroad. Within a research project the question arose about the reasons why a constant growing number of U.S. citizens are expatriating – the answers are believed helping to explain the underlying governmental policy rational, leading to such activities. While it is impossible to locate former U.S. citizens to conduct a survey on the reasons and the U.S. government is not commenting on the reasons given within the process of expatriation, the chosen methodology is Structural Equation Modeling (SEM), in the first step by re-using current surveys conducted by different researchers within the population of U. S. citizens residing abroad during the last years. Surveys questioning the personal situation in the context of tax, compliance, citizenship and likelihood to repatriate to the U. S. In general SEM allows: (1) Representing, estimating and validating a theoretical model with linear (unidirectional or not) relationships. (2) Modeling causal relationships between multiple predictors (exogenous) and multiple dependent variables (endogenous). (3) Including unobservable latent variables. (4) Modeling measurement error: the degree to which observable variables describe latent variables. Moreover SEM seems very appealing since the results can be represented either by matrix equations or graphically. Results: the observed variables (items) of the construct are caused by various latent variables. The given surveys delivered a high correlation and it is therefore impossible to identify the distinct effect of each indicator on the latent variable – which was one desired result. Since every SEM comprises two parts: (1) measurement model (outer model) and (2) structural model (inner model), it seems necessary to extend the given data by conducting additional research and surveys to validate the outer model to gain the desired results.

Keywords: expatriation of U. S. citizens, SEM, structural equation modeling, validating

Procedia PDF Downloads 196

Authors: Azeddine El-Hassouny, Chandrashekhar Meshram, Geraldin Nanfack

Abstract:

In recent years, learning from data with noisy labels (Label Noise) has been a major concern in supervised learning. This problem has become even more worrying in Deep Learning, where the generalization capabilities have been questioned lately. Indeed, deep learning requires a large amount of data that is generally collected by search engines, which frequently return data with unreliable labels. In this paper, we investigate the Label Noise in Deep Learning using variational inference. Our contributions are : (1) exploiting Label Noise concept where the true labels are learnt using reparameterization variational inference, while observed labels are learnt discriminatively. (2) the noise transition matrix is learnt during the training without any particular process, neither heuristic nor preliminary phases. The theoretical results shows how true label distribution can be learned by variational inference in any discriminate neural network, and the effectiveness of our approach is proved in several target datasets, such as MNIST and CIFAR32.

Keywords: label noise, deep learning, discrete latent variable, variational inference, MNIST, CIFAR32

Procedia PDF Downloads 107

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

Abstract:

In this paper, we propose a novel inference algorithm for the multi-class Gaussian process classification model that can be used in the field of human behavior recognition. This algorithm can drive simultaneously both a posterior distribution of a latent function and estimators of hyper-parameters in a Gaussian process classification model with multi-class. Our algorithm is based on the Laplace approximation (LA) technique and variational EM framework. This is performed in two steps: called expectation and maximization steps. First, in the expectation step, using the Bayesian formula and LA technique, we derive approximately the posterior distribution of the latent function indicating the possibility that each observation belongs to a certain class in the Gaussian process classification model. Second, in the maximization step, using a derived posterior distribution of latent function, we compute the maximum likelihood estimator for hyper-parameters of a covariance matrix necessary to define prior distribution for latent function. These two steps iteratively repeat until a convergence condition satisfies. Moreover, we apply the proposed algorithm with human action classification problem using a public database, namely, the KTH human action data set. Experimental results reveal that the proposed algorithm shows good performance on this data set.

Keywords: bayesian rule, gaussian process classification model with multiclass, gaussian process prior, human action classification, laplace approximation, variational EM algorithm

Procedia PDF Downloads 316

Authors: Mariusz Hofman

Abstract:

The paper ‘Theoretical framework for value creation in Project-Oriented Companies’ is designed to determine, how organisations create value and whether this allows them to achieve market success. An assumption has been made that there are two routes to achieving this value. The first one is to create intangible assets (i.e. the resources of human, structural and relational capital), while the other one is to create added value (understood as the surplus of revenue over costs). It has also been assumed that the combination of the achieved added value and unique intangible assets translates to the success of a project-oriented company. The purpose of the paper is to present hypothetical and deductive model which describing the modus operandi of such companies and approach to model operationalisation. All the latent variables included in the model are theoretical constructs with observational indicators (measures). The existence of a latent variable (construct) and also submodels will be confirmed based on a covariance matrix which in turn is based on empirical data, being a set of observational indicators (measures). This will be achieved with a confirmatory factor analysis (CFA). Due to this statistical procedure, it will be verified whether the matrix arising from the adopted theoretical model differs statistically from the empirical matrix of covariance arising from the system of equations. The fit of the model with the empirical data will be evaluated using χ2, RMSEA and CFI (Comparative Fit Index). How well the theoretical model fits the empirical data is assessed through a number of indicators. If the theoretical conjectures are confirmed, an interesting development path can be defined for project-oriented companies. This will let such organisations perform efficiently in the face of the growing competition and pressure on innovation.

Keywords: value creation, project-oriented company, structural equation modelling

Procedia PDF Downloads 263

Authors: Elnaz Saeedi, Jamileh Abolaghasemi, Mohsen Nasiri Tousi, Saeedeh Khosravi

Abstract:

Goals and Objectives: A typical analysis of survival data involves the modeling of time-to-event data, such as the time till death. A frailty model is a random effect model for time-to-event data, where the random effect has a multiplicative influence on the baseline hazard function. This article aims to investigate the use of gamma frailty model with concomitant variable in order to individualize the prognostic factors that influence the liver cirrhosis patients’ survival times. Methods: During the one-year study period (May 2008-May 2009), data have been used from the recorded information of patients with liver cirrhosis who were scheduled for liver transplantation and were followed up for at least seven years in Imam Khomeini Hospital in Iran. In order to determine the effective factors for cirrhotic patients’ survival in the presence of latent variables, the gamma frailty distribution has been applied. In this article, it was considering the parametric model, such as Exponential and Weibull distributions for survival time. Data analysis is performed using R software, and the error level of 0.05 was considered for all tests. Results: 305 patients with liver cirrhosis including 180 (59%) men and 125 (41%) women were studied. The age average of patients was 39.8 years. At the end of the study, 82 (26%) patients died, among them 48 (58%) were men and 34 (42%) women. The main cause of liver cirrhosis was found hepatitis 'B' with 23%, followed by cryptogenic with 22.6% were identified as the second factor. Generally, 7-year’s survival was 28.44 months, for dead patients and for censoring was 19.33 and 31.79 months, respectively. Using multi-parametric survival models of progressive and regressive, Exponential and Weibull models with regard to the gamma frailty distribution were fitted to the cirrhosis data. In both models, factors including, age, bilirubin serum, albumin serum, and encephalopathy had a significant effect on survival time of cirrhotic patients. Conclusion: To investigate the effective factors for the time of patients’ death with liver cirrhosis in the presence of latent variables, gamma frailty model with parametric distributions seems desirable.

Keywords: frailty model, latent variables, liver cirrhosis, parametric distribution

Procedia PDF Downloads 247

Authors: Suphap Sikkhaphan, Muwanga Zake, Johnnie Wycliffe Frank

Abstract:

The purposes of the study included a) investigating the authentic and transformational leadership model of the directors of tambon health promoting hospitals b) evaluating the relation between the authentic and transformation leadership of the directors of tambon health promoting hospitals and the effectiveness of their hospitals and c) assessing the invariance test of the authentic and transformation leadership of the directors of tambon health promoting hospitals. All 400 southern tambon health promoting hospital directors were enrolled into the study. Half were males (200), and another half were females (200). They were sampled via a stratified method. A research tool was a questionnaire paper containing 4 different sections. The Alpha-Cronbach’s Coefficient was equally to .98. Descriptive analysis was used for demographic data, and inferential statistics was used for the relation and invariance tests of authentic and transformational leadership of the directors of tambon health promoting hospitals. The findings revealed overall the authentic and transformation leadership model of the directors of tambon health promoting hospitals has the relation to the effectiveness of the hospitals. Only the factor of “strong community support” was statistically significantly related to the authentic leadership (p < .05). However, there were four latent variables statistically related to the transformational leadership including, competency and work climate, management system, network cooperation, and strong community support (p = .01). Regarding the relation between the authentic and transformation leadership of the directors of tambon health promoting hospitals and the effectiveness of their hospitals, four casual variables of authentic leadership were not related to those latent variables. In contrast, all four latent variables of transformational leadership has statistically significantly related to the effectiveness of tambon health promoting hospitals (p = .001). Furthermore, only management system variable was significantly related to those casual variables of the authentic leadership (p < .05). Regarding the invariance test, the result found no statistical significance of the authentic and transformational leadership model of the directors of tambon health promoting hospitals, especially between male and female genders (p > .05).

Keywords: authentic leadership, transformational leadership, tambon health promoting hospital

Procedia PDF Downloads 420

Authors: Gyoomi Kim, Jiyoung Bae

Abstract:

The purpose of this study was to propose a framework that verifies the structural relationships among students’ use of digital English learning strategy (DELS), affective domains, and their individual variables. The study developed a hypothetical model based on previous studies on language learning strategy use as well as digital language learning. The participants were 720 Korean high school students and 430 university students. The instrument was a self-response questionnaire that contained 70 question items based on Oxford’s SILL (Strategy Inventory for Language Learning) as well as the previous studies on language learning strategies in digital learning environment in order to measure DELS and affective domains. The collected data were analyzed through structural equation modeling (SEM). This study used quantitative data analysis procedures: Explanatory factor analysis (EFA) and confirmatory factor analysis (CFA). Firstly, the EFA was conducted in order to verify the hypothetical model; the factor analysis was conducted preferentially to identify the underlying relationships between measured variables of DELS and the affective domain in the EFA process. The hypothetical model was established with six indicators of learning strategies (memory, cognitive, compensation, metacognitive, affective, and social strategies) under the latent variable of the use of DELS. In addition, the model included four indicators (self-confidence, interests, self-regulation, and attitude toward digital learning) under the latent variable of learners’ affective domain. Secondly, the CFA was used to determine the suitability of data and research models, so all data from the present study was used to assess model fits. Lastly, the model also included individual learner factors as covariates and five constructs selected were learners’ gender, the level of English proficiency, the duration of English learning, the period of using digital devices, and previous experience of digital English learning. The results verified from SEM analysis proposed a theoretical model that showed the structural relationships between Korean students’ use of DELS and their affective domains. Therefore, the results of this study help ESL/EFL teachers understand how learners use and develop appropriate learning strategies in digital learning contexts. The pedagogical implication and suggestions for the further study will be also presented.

Keywords: Digital English Learning Strategy, DELS, individual variables, learners' affective domains, Structural Equation Modeling, SEM

Procedia PDF Downloads 110

Authors: Ehsan Motamedian

Abstract:

Finding all optimal flux distributions of a metabolic model is an important challenge in systems biology. In this paper, a new algorithm is introduced to identify all alternate optimal solutions of a large scale metabolic network. The algorithm reduces the model to decrease computations for finding optimal solutions. The algorithm was implemented on the Escherichia coli metabolic model to find all optimal solutions for lactate and acetate production. There were more optimal flux distributions when acetate production was optimized. The model was reduced from 1076 to 80 variable fluxes for lactate while it was reduced to 91 variable fluxes for acetate. These 11 more variable fluxes resulted in about three times more optimal flux distributions. Variable fluxes were from 12 various metabolic pathways and most of them belonged to nucleotide salvage and extra cellular transport pathways.

Keywords: flux variability, metabolic network, mixed-integer linear programming, multiple optimal solutions

Procedia PDF Downloads 418

Authors: Rachid Belhachemi, Pierre Rostan, Alexandra Rostan

Abstract:

This paper develops a discrete-time option pricing model for index options. The model consists of two key ingredients. First, daily stock return innovations are driven by a continuous hidden threshold mixed skew-normal (HTSN) distribution which generates conditional non-normality that is needed to fit daily index return. The most important feature of the HTSN is the inclusion of a latent state variable with a continuum of states, unlike the traditional mixture distributions where the state variable is discrete with little number of states. The HTSN distribution belongs to the class of univariate probability distributions where parameters of the distribution capture the dependence between the variable of interest and the continuous latent state variable (the regime). The distribution has an interpretation in terms of a mixture distribution with time-varying mixing probabilities. It has been shown empirically that this distribution outperforms its main competitor, the mixed normal (MN) distribution, in terms of capturing the stylized facts known for stock returns, namely, volatility clustering, leverage effect, skewness, kurtosis and regime dependence. Second, heteroscedasticity in the model is captured by a threeexogenous-factor GARCH model (GARCHX), where the factors are taken from the principal components analysis of various world indices and presents an application to option pricing. The factors of the GARCHX model are extracted from a matrix of world indices applying principal component analysis (PCA). The empirically determined factors are uncorrelated and represent truly different common components driving the returns. Both factors and the eight parameters inherent to the HTSN distribution aim at capturing the impact of the state of the economy on price levels since distribution parameters have economic interpretations in terms of conditional volatilities and correlations of the returns with the hidden continuous state. The PCA identifies statistically independent factors affecting the random evolution of a given pool of assets -in our paper a pool of international stock indices- and sorting them by order of relative importance. The PCA computes a historical cross asset covariance matrix and identifies principal components representing independent factors. In our paper, factors are used to calibrate the HTSN-GARCHX model and are ultimately responsible for the nature of the distribution of random variables being generated. We benchmark our model to the MN-GARCHX model following the same PCA methodology and the standard Black-Scholes model. We show that our model outperforms the benchmark in terms of RMSE in dollar losses for put and call options, which in turn outperforms the analytical Black-Scholes by capturing the stylized facts known for index returns, namely, volatility clustering, leverage effect, skewness, kurtosis and regime dependence.

Keywords: continuous hidden threshold, factor models, GARCHX models, option pricing, risk-premium

Procedia PDF Downloads 288

Authors: Mohsen Farhadloo, Majid Farhadloo

Abstract:

Aspect-level opinion mining that aims at discovering aspects (aspect identification) and their corresponding ratings (sentiment identification) from customer reviews have increasingly attracted attention of researchers and practitioners as it provides valuable insights about products/services from customer's points of view. Instead of addressing aspect identification and sentiment identification in two separate steps, it is possible to simultaneously identify both aspects and sentiments. In recent years many graphical models based on Latent Dirichlet Allocation (LDA) have been proposed to solve both aspect and sentiment identifications in a single step. Although LDA models have been effective tools for the statistical analysis of document collections, they also have shortcomings in addressing some unique characteristics of opinion mining. Our goal in this paper is to address one of the limitations of topic models to date; that is, they fail to directly model the associations among topics. Indeed in many text corpora, it is natural to expect that subsets of the latent topics have higher probabilities. We propose a probabilistic graphical model called focus-LDA, to better capture the associations among topics when applied to aspect-level opinion mining. Our experiments on real-life data sets demonstrate the improved effectiveness of the focus-LDA model in terms of the accuracy of the predictive distributions over held out documents. Furthermore, we demonstrate qualitatively that the focus-LDA topic model provides a natural way of visualizing and exploring unstructured collection of textual data.

Keywords: aspect-level opinion mining, document modeling, Latent Dirichlet Allocation, LDA, sentiment analysis

Procedia PDF Downloads 81

Authors: Shuaibu Hassan Usman, Ishaq Oyebisi Oyefolahan

Abstract:

This study was aimed to explore the latent factors in the web 2.0 enabled knowledge sharing practices instrument. Seven latent factors were identified through a factor analysis with orthogonal rotation and interpreted based on simple structure convergence, item loadings, and analytical statistics. The number of factors retains was based on the analysis of Kaiser Normalization criteria and Scree plot. The reliability tests revealed a satisfactory reliability scores on each of the seven latent factors of the web 2.0 enabled knowledge sharing practices. Limitation, conclusion, and future work of this study were also discussed.

Keywords: factor analysis, latent factors, knowledge sharing practices, students, web 2.0 enabled

Procedia PDF Downloads 420

Authors: Pontus Backstrom

Abstract:

In the educational literature on peer effects, attention has been brought to the fact that the mechanisms creating peer effects are still to a large extent hidden in obscurity. The hypothesis in this study is that the Frame Factor Theory can be used to explain these mechanisms. At heart of the theory is the concept of “time needed” for students to learn a certain curricula unit. The relations between class-aggregated time needed and the actual time available, steers and hinders the actions possible for the teacher. Further, the theory predicts that the timing and pacing of the teachers’ instruction is governed by a “criterion steering group” (CSG), namely the pupils in the 10th-25th percentile of the aptitude distribution in class. The class composition hereby set the possibilities and limitations for instruction, creating peer effects on individual outcomes. To test if the theory can be applied to the issue of peer effects, the study employs multilevel structural equation modelling (M-SEM) on Swedish TIMSS 2015-data (Trends in International Mathematics and Science Study; students N=4090, teachers N=200). Using confirmatory factor analysis (CFA) in the SEM-framework in MPLUS, latent variables are specified according to the theory, such as “limitations of instruction” from TIMSS survey items. The results indicate a good model fit to data of the measurement model. Research is still in progress, but preliminary results from initial M-SEM-models verify a strong relation between the mean level of the CSG and the latent variable of limitations on instruction, a variable which in turn have a great impact on individual students’ test results. Further analysis is required, but so far the analysis indicates a confirmation of the predictions derived from the frame factor theory and reveals that one of the important mechanisms creating peer effects in student outcomes is the effect the class composition has upon the teachers’ instruction in class.

Keywords: compositional effects, frame factor theory, peer effects, structural equation modelling

Procedia PDF Downloads 120

Authors: Narong Kulnides

Abstract:

This research involved fingerprints on ballistics after shooting. Two objectives of research were as follows; (1) to study the duration of the existence of latent fingerprints on .38, .45, 9 mm and .223 cartridge case after shooting, and (2) to compare the effectiveness of the detection of latent fingerprints by Black Powder, Super Glue, Perma Blue and Gun Bluing. The latent fingerprint appearance were studied on .38, .45, 9 mm. and .223 cartridge cases before and after shooting with Black Powder, Super Glue, Perma Blue and Gun Bluing. The detection times were 3 minute, 6, 12, 18, 24, 30, 36, 42, 48, 54, 60, 66, 72, 78 and 84 hours respectively. As a result of the study, it can be conclude that: (1) Before shooting, the detection of latent fingerprints on 38, .45, and 9 mm. and .223 cartridge cases with Black Powder, Super Glue, Perma Blue and Gun Bluing can detect the fingerprints at all detection times. (2) After shooting, the detection of latent fingerprints on .38, .45, 9 mm. and .223 cartridge cases with Black Powder, Super Glue did not appear. The detection of latent fingerprints on .38, .45, 9 mm. cartridge cases with Perma Blue and Gun Bluing were found 100% of the time and the detection of latent fingerprints on .223 cartridge cases with Perma Blue and Gun Bluing were found 40% and 46.67% of the time, respectively.

Keywords: ballistic, fingerprint, shooting, detection times

Procedia PDF Downloads 406

Authors: Fujio Akagi, Hiroaki Ito, Shin-Ichi Inage

Abstract:

The aim of this study is to develop a combustion model that can be applied uniformly to laminar and turbulent premixed flames while considering the effect of the Lewis number (Le). The model considers the effect of Le on the transport equations of the reaction progress, which varies with the chemical species and temperature. The distribution of the reaction progress variable is approximated by a hyperbolic tangent function, while the other distribution of the reaction progress variable is estimated using the approximated distribution and transport equation of the reaction progress variable considering the Le. The validity of the model was evaluated under the conditions of propane with Le > 1 and methane with Le = 1 (equivalence ratios of 0.5 and 1). The estimated results were found to be in good agreement with those of previous studies under all conditions. A method of introducing a turbulence model into this model is also described. It was confirmed that conventional turbulence models can be expressed as an approximate theory of this model in a unified manner.

Keywords: combustion model, laminar flame, Lewis number, turbulent flame

Procedia PDF Downloads 109

Authors: Gabriel Calvo, Carmen Armero, Luigi Spezia

Abstract:

A basketball player is said to have a hot hand if his/her performance is better than expected in different periods of time. A way to deal with this phenomenon is to make use of latent variables, which can indicate whether the player is ‘on fire’ or not. This work aims to model the hot hand phenomenon through a Bayesian hidden Markov model (HMM) with two states (cold and hot) and two different probability of success depending on the corresponding hidden state. This task is illustrated through a comprehensive simulation study. The simulated data sets emulate the field goal attempts in an NBA season from different profile players. This model can be a powerful tool to assess the ‘streakiness’ of each player, and it provides information about the general performance of the players during the match. Finally, the Bayesian HMM allows computing the posterior probability of any type of streak.

Keywords: Bernoulli trials, field goals, latent variables, posterior distribution

Procedia PDF Downloads 170

Authors: Kaouther Selmi, Alaeddine Sridi, Mohamed Bouallegue, Kais Bouallegue

Abstract:

Neurons are the fundamental units of the brain and the nervous system. The variable structure model of neurons consists of a system of differential equations with various parameters. By optimizing these parameters, we can create a unique model that describes the dynamic behavior of a single neuron. We introduce a neural network based on neurons with multiple dendrites employing an activation function with a variable structure. In this paper, we present a model for heart behavior. Finally, we showcase our successful simulation of the heart's ECG diagram using our Variable Structure Neuron Model (VSMN). This result could provide valuable insights into cardiology.

Keywords: neural networks, neuron, dendrites, heart behavior, ECG

Procedia PDF Downloads 64

Authors: Murat Gunduz, Mustafa Ozdemir

Abstract:

In this research, a framework is to be proposed to model the safety performance in construction sites. Determinants of safety performance are to be defined through extensive literature review and a multidimensional safety performance model is to be developed. In this context, a questionnaire is to be administered to construction companies with sites. The collected data through questionnaires including linguistic terms are then to be defuzzified to get concrete numbers by using fuzzy set theory which provides strong and significant instruments for the measurement of ambiguities and provides the opportunity to meaningfully represent concepts expressed in the natural language. The validity of the proposed safety performance model, relationships between determinants of safety performance are to be analyzed using the structural equation modeling (SEM) which is a highly strong multi variable analysis technique that makes possible the evaluation of latent structures. After validation of the model, a safety performance index assessment tool is to be proposed by the help of software. The proposed safety performance assessment tool will be based on the empirically validated theoretical model.

Keywords: Fuzzy set theory, safety performance assessment, safety index, structural equation modeling (SEM), construction sites

Procedia PDF Downloads 497