"Annals. Computer Science Series" Journal Romania, 300559 Timişoara, 6 Lascăr Catargiu str. Phone: 004 0256 220 687 E-mail: conference.fcia [@] tibiscus [.] ro |
|
|
|
» Topp Leone Exponential – exponential distributions properties and application
» A. A. Sanusi, S.I.S Doguwa, I. Audu, Y.M. Baraya ABSTRACT: In this paper, a new distribution called ToppLeone Exponential – Exponential distribution (TLEE) is developed by making Topp Leone Exponential G family as Generator while exponential distribution as the base line distribution. The respective density and distribution functions of this distribution were shown. Some structural properties of this new distribution were derived such as moment generating function, renyi entropy and order statistics.Maximum Likelihood Estimate (MLE) methods was used to estimate the parameters of this new distribution. Finally, two real data sets were used to validate the results obtained from MLE. The results shown that TLEE distribution provide better fit in the data sets than some other commonly known distributions. Perhaps, this new distribution may be used to modeling positive real life data sets. KEYWORDS:Topp Leone Exponential – Exponential distribution, Moment generating function, Renyi entropy, Order statistics, maximum likelihood estimation |
9 - 14
|
|
|
» Bayesian Estimation of Dynamic Error Component Model with Moving Average Disturbances » Adepoju Abosede Adedayo and Ebukuyo Oluremi Bolaji ABSTRACT: This paper explores the estimation of Dynamic Panel Data Error Component Model with Autocorrelated disturbance of Moving Average of order one MA(1) using Bayesian approach; the Bayesian approach produced a very efficient estimator that was not plagued by moment conditions as in other popular Generalized method of moments (GMM) estimators. The Markov Chain Monte Carlo (MCMC) with Gibbs sampler algorithm was used to carry out the analysis as shown in the empirical illustrations. The estimation was done at varying degrees of Autocorrelation coefficients that is from mild to severe. The result showed that the Bayesian Estimator performed excellently as the Numerical Standard Error (NSE) decreased with increasing sample sizes. KEYWORDS:Dynamic Panel Data, Error Component, Autocorrelated Disturbance, Moving Average, Markov Chain Monte Carlo (MCMC), Gibbs Sampler. |
15 - 20
|
|
|
» Online Hospital Management System » Muhsin Hassanu Saleh and Mukhtar Bello ABSTRACT: This paper aims to address the online hospital management system as an area to enhance medical field. Patients tends to spend a lot of time for treatment, this tends to result in patient being attended lately, the manual system was looked in detail with a help to find out the need to improved the system. A computerized program was designed to bring about improvement in the care of individual patients, with a view of taking the advantage of computer speed, storage and retrieved facilities. The software was designed to take care of patient’s registration, billing, treatment and payments. KEYWORDS:Hospital, management system. |
21 - 27
|
|
|
» Polynomial Sequence and Exponential Generating Function of Lucas Series » Tran Trung Tinh and Dam Van Nhi ABSTRACT: In teaching mathematics at high schools, the mathematics competency of teacher is an important procedure in teaching process. In this paper we study the properties of polynomial sequence and exponential generating function of Lucas series and I pointed out some application in mathematics. And I hope it will help the teachers to develop high school their mathematics competency in Vietnam. KEYWORDS:Teachers' mathematics competency, Polynomial sequence, Exponential generating functions, Lucas series. |
28 - 33
|
|
|
» Computing Fourier Integral using Matlab » A. R.Tasiu and H. Usman ABSTRACT: This work studies the computation of Fourier integral using Matlab software. The expressions obtained were computed in Matlab using ezplot command. The plots in line and contour form were presented. From the outcome of the result it was observed that the functions increase with increasing values of π. KEYWORDS:Fourier integral, matlab. |
34 - 38
|
|
|
» Cointegration Analysis on Unemployment and Macroeconomic Variables among selected African Countries (A Dynamic Panel Approach) » Ayoola F. J. and Emmanuel C. Ezeja ABSTRACT: Unemployment in Nigeria and the entire sub-Saharan countries is viewed as part of major problem hindering development in Africa. Unemployment and its determinant have been widely discussed in literatures and have continued to attract the attention of researchers especially applied econometricians. The constant increase of unemployment in Africa necessitated this study. This study examined the effect of some selected macroeconomic variable on unemployment among 26 African countries using cointegration analysis. The panel unit root test shows that the entire variable contains a unit root at the level stage, while the first differences of the variables are all stationary. The result of Johansen cointegration test revealed that there is a long run relationship which exists between the selected macroeconomic variables and unemployment. We therefore employed the dynamic ordinary least square (DOLS) and fully modified ordinary least square (FMOLS) to obtaining the coefficient of the long run. The result of the DOLS indicates that at 5% level of significance the gross domestic product and the foreign direct investment have a negative and positive impact on unemployment respectively while exchange rate is not significant at the long run. Similarly the result of the FMOLS is in consistence with the DOLS. However we will rely on the DOLS for our inference since it performs better than FMOLS according to recent literatures. KEYWORDS:Panel unit root, Johansen Cointegration test, dynamic ordinary least square, dynamic panel models, fully modified ordinary least square(FMOLS). |
39 - 46
|
|
|
» Estimation of the Population Mean in the Presence of Measurement and Response Error » Olayiwola O. M, Ajayi A. O., Akintunde A. A., Wale-Orojo O. A., Ajibade F. B. and Ojelabi O. ABSTRACT: Call back helps to reduce non-response in survey. Existing works had investigated the problem of errors in survey, but there is little or no of information in existing work where these problems exist together in survey. This work derived an estimator which incorporates the problem of non-response and measurement error. The data considered is the height and weight of students with weight as an auxiliary variable. Descriptive statistics were used to determine average height and weight while box plot was used to check for outliers across the height and weight. The average height and weight for the students were 4.488 feet and 53.95kg respectively, while the box plot indicated that there were no outliers. The Mean Square error (MSE) of the proposed estimator when there is non-response error alone was 20.5454, MSE of the proposed estimator when there is measurement error alone was 29.2543, the MSE of the new estimator when there is non-response and measurement error was gotten to be 73.3735 while for the existing es was gotten to be 206.551. It can be noticed that the MSE of the proposed estimator was lesser than the existing, hence more efficient. KEYWORDS:Mean Square error (MSE), estimator, measurement error |
47 - 51
|
|
|
» An efficient automated revenue generation database management system » D. D. Wisdom, E. A. Ajayi, S. O. Akindayo, Y. M. Yanah, E. Kwaido and S. A. Shehu ABSTRACT: Missing documents, multiple record entries, redundancy in data entry and revenue records over the years are a critical challenge. Thus, we have proposed An Efficient Automated Revenue Generation Database management system. The software is designed to maintain accurate and relevant revenue generated records efficiently. The research study tries to mitigate the problem of mismanagement of revenue records, multiple record entries, missing records, redundancy of records as well as effective management of revenue records/live storage of these records. The research also established a good user oriented data bank and designed a suitable database, procuring and installing the necessary database, communication and application software and developed a detailed documentation of all revenue allocated or collected which will encourage accountability and transparency effectively. KEYWORDS:Automation, Database Management System, Revenue Records |
52 - 64
|
|
|
» A time series modelling of environmental degradation as its affects mortality » Oyedeji, Marufat Bukola ABSTRACT: The relationship between air pollution and mortality has gained increasing attention in the past decade. Many time-series analyses have been conducted worldwide, including in all the major cities in the Africa, United States, Europe, and Asia. This conducted time-series analysis of Oyo State of Nigeria and then use a time series model to pool the results and construct a dose-response relationship and generate a predictive model. The study analyses the effect of environmental pollution on mortality in Oyo State, Nigeria. Data from the study were analyses using the Vector Autoregressive (VAR) Model. The study revealed that environmental pollution has impact on mortality in Oyo State, Nigeria. In conclusion, the study recommended a collaborative policy by government, and other stakeholders on curb diseases in oil producing areas. KEYWORDS:Air pollution, vector autoregressive, modelling |
65 - 69
|
|
|
» Regression techniques in the presence of multicollinearity and autocorrelation phenomena: Monte Carlo approach » O. Oyewole, D.A. Agunbiade ABSTRACT: Multicollinearity and Autocorrelation are two very common problems in regression analysis. As its well-known, the presence of some degrees of multicollinearity results in estimation instability and model mis-specification while the presence of serial correlated errors lead to underestimation of the variance of parameter estimates and inefficient prediction. These two conditions have adverse effects on estimation and prediction; therefore, a wide range of tests have been developed to reduce their impact. Invariably, the multicollinearity and autocorrelation problems are dealt with separately in most studies. Thus, this study explored the predictive ability of the proposed GLS-Ridge regression on multicollinearity and autocorrelation problems simultaneously, using simulated dataset. Data used for the study was the data simulated using Monte Carlo. The research work revealed that the GLS-R regression technique has a better predictive ability in the presence of autocorrelation and multicollinearity, hence it is preferred than the other three techniques. KEYWORDS:Ridge regression, lasso, Monte Carlo simulation, multicollinearity, autocorrelation |
70 - 77
|
|
|
» Analysis of family planning as a predictor of women lifespan » Oyedeji, Marufat Bukola ABSTRACT: The life expectancy for women in the United States is 77.34 years with women who engage in family planning by taking oral contraceptives for 5 years before the age of 30 years are most likely to live additional 4 days more. This study examines the women lifespan is predicted by family planning with data collected from University College Hospital, Ibadan and Lagos Bureau of Statistics and World Bank Statistical database. The study employed regression model to analyse the data. Augmented Dickey-Fuller test used to test for stationarity, which showed that the data are stationary. The result of the study indicates that there is significant effect of women lifespan predicted method of contraceptives and age. The R-square value of 0.9337 showed that about 93.37% of the total variation in life expectancy is accounted for by the use of contraceptive (injectable, Oral, pills& IUCD) and age with other variables in the expectancy. The study therefore recommended among others the need to provide family planning education, so that people can understand the use, cope with it, take control and develop survival skills. KEYWORDS:family planning, lifespan, regression model, R-square |
78 - 84
|
|
|
» Modification of ratio type estimator for population mean » K.A Muhammed, L. Adebara, I.O.Ajao ABSTRACT: Ratio estimator is one of the estimators that make use of auxiliary information to improve the precisions of the population mean. Many researchers have been proposed or modified ratio-type estimators in different techniques. In this research work, ratio-type estimator has been proposed under simple random sampling without replacement, the Bias and Mean Square Error of the proposed estimator have been obtained up to first order approximation using Taylor’s Series Expansions as a properties of proposed estimator. The conditions for its efficiency over some existing estimators have been established. The proposed estimator exhibit significant improvement over conversional and existing ratio- type estimators considered in this study, this was discovered as a result of optimum value of their constant that leads to minimum value of its Mean Square Error compared to existing estimators in this study. Two populations were considered to demonstrate the performances of the estimators at the end of the study it was discovered that, the Bias and Mean Square Error of the proposed estimator minimized compared to existing estimators. Therefore, the proposed estimator performed better than existing estimators considered in this study. KEYWORDS:Auxiliary Variable, Efficiency, Bias and Mean Square Error |
85 - 88
|
|
|
» Blood flow through a multi-stenosed human carotid artery: effects of hematocrit variation and stenosis number » Sefiu Adekunle Onitilo ABSTRACT: Stenosis is the development of plague in the human arteries that hinders the flow of blood in the arteries. Also, the deposits of cholesterol on the arterial wall and proliferation of connective tissues are responsible for the abnormal growth in the lumen of an artery. This paper investigates the flow of blood through a multi-stenosed human carotid artery, considering the effect of Hematocrit and stenosis variation. In this paper, it was discovered that increase in the number of stenosis, fluid characteristics (the wall shear stress, pressure gradient, viscosity of the blood) and increase in Hematocrit (H) of red blood cells is dangerous to human heart. This can result into bleeding and blood clotting of the vessels which can break the human heart, and result into heart attack or stroke. KEYWORDS:stenosis, blood flow, wall shear stress, hematocrit, blood viscosity |
89 - 94
|
|
|
» Preventing shoulder surfing attack in graphical password authentication scheme » Hammed, M., Adeboye, N. O. ABSTRACT: Authentication based password is very common in computer security and privacy. Most of the traditional passwords are numbers, numbers with alphabets and numbers with alphabets and symbols. That can be easily broken by the attacks such as eves dropping, dictionary attacks, social engineering and shoulder surfing attacks. However, human factors such as choosing bad passwords and keeping passwords in an insecure place are also big problems. In order to address these challenges, several graphical authentication schemes have been proposed, but still shoulder surfing attack increasing. This study proposed multiplication matrix for graphical authentication scheme to reduce shoulder surfing attack. The system attained high degree of accuracy to restrict unauthorized users. KEYWORDS:Graphical authentication, shoulder surfing attack, Multiplication matrix, Password |
95 - 98
|
|
|
» Modelling the fetus mortality due to abortion among Nigeria women: a binary response analysis » Olubimpe M. Oladuti ABSTRACT: Termination of pregnancy occur due to sexual laissez-faire of the society in particular premarital sex, pressure from a sex partner not to use a contraceptive, poverty and low socio-economic status. The aim of the study was to investigate with the application of binary logit model the risk factors associated with the prevalence of abortion among Nigerian women on data extracted from 2013 Nigeria Demographic and Health Survey (NDHS). The results of the analysis shows that fetus mortality as a result of abortion depend on geo-political zone, location, level of education attainment, religion, forced sexual acts, total children ever born, age at first sexual initiation. It was also found that lack of knowledge about contraceptives use; household wealth index, ever heard of HIV/AIDS and being employed were associated with increased likelihood of abortion among Nigeria women. KEYWORDS:Pregnancy, Fetus mortality, Abortion, Risk factors, Binary logit, Nigeria, Demographic and Health Survey |
99 - 105
|
|
|
» Global stability of lymphatic filariasis in Ekiti State Nigeria » Bukola Badeji-Ajisafe ABSTRACT: Lymphatic filariasis (LF) is one of a major tropical disease in Nigeria most especially in Ekiti State, it is highly prevalence and severe among the rural dwellers. The disease affects majorly people in the rural areas where there is an abject poverty, LF is a mosquito-borne disease endemic in tropical regions and caused by the parasitic nematode Wuchereria bancrofti in Africa, and in Southeast Asia. A mathematical model called SEIR (Susceptible, Exposed, Infectious and Recovered) is used for the analysis of LF. The stability analysis of the Disease-Free Equilibria (DFE) and Endemic Equilibria (EE) were measured to know the prevalence level of the disease in Ekiti State. Also, the equilibrium level were determined to know if the models were locally asymptotically stable or unstable, when the reproduction number is less than one (R0 < 1) or when is greater than one (R0 >1). Global stability of (DFE) and (EE) was also considered in the model, using Lassalle’s invariance principle of Lyapunov functions. KEYWORDS:Global Stability, lymphatic Filariasis, Mathematical modelling, epidemiological model, SEIR |
106 - 111
|
|
|
» Design and implementation of an improved realtime food and drug authenticating and interacting system
(a case study of NAFDAC)
» Ndunagu, J. N., Chuks-Ugochukwu Chinenye M. ABSTRACT: The problem of fake products by counterfeiters and unregistered manufacturers is real and constitutes a major threat to the health and safety of the Nigerian population. Fake drugs are a major issue in Nigeria with thousands losing their lives every year to fake and adulterated drug in the country. The Mobile Service for Food and Drug Authentication is a system that ascertains the validity of drugs and foods using mobile phones. The National Agency for Food and Drug Administration and Control (NAFDAC) already has a Mobile Authentication Service (MAS) that enables consumers verify few drugs with SMS and USSD, but this current system is faced with usability issues (as there are indications that the present system has limitation to achieve 100% authentication process since the system lacks sufficient interfaces and can be copied by counterfeiters) which the proposed system aims to fix. Using the Dynamic System Development Methodology (DSDM), a more technically robust, cost-effective and portable real-time device was created, so that during a query session, the C program codes on the embedded system triggers the SIM808 to make an internet connection to the central database (MySQL server) containing valid products data, in order to confirm products authenticities. This new device thus re-formatted the existing system user interfaces in other to make it easier for users to interact with the system and hereby creating a device (that can be singly manned by NAFDAC instead of outsourced to third party agents) to validate products with ease and as well enable prompts for the report of invalid products which where all the former lacks. KEYWORDS:Authentic Product, Consumer Verification System, Counterfeit Product, NAFDAC |
112 - 117
|
|
|
» Feature extraction for sms spam detection: review » Taofeek-Ibrahim Fatimoh Abidemi, Toye Nike Toyin ABSTRACT: The general acceptance of mobile phones has resulted to a noticeable increase in sending of Short Message Service (SMS) messages. Messages through SMS have been considered as a rapid method of communication due to its stress-free approach and low cost, as a result, SMS has been targeted for several forms of threats, one of these is spamming. The SMS spam is unsolicited or undesirable message sent to many mobile phone users, and cause many problems like displeasure, consumption of network bandwidth, scam and theft of personal information and malware installation. Feature extraction is a significant phase of SMS Spam detection, the extraction of features in SMS involves a process of reduction of an initial set of raw features into to more manageable forms for classification process. This paper conducted a literature review on some of the commonly used feature extraction techniques for SMS Spam detection system. A contribution was made on how these techniques can be further improved in order to develop robust SMS Spam detection system. KEYWORDS:Feature Extraction, Network Bandwidth, Short Message Service, Spam SMS |
118 - 121
|
|
|
» Deep learning based real – time facial recognition system for identity management » Oyeniran, Oluwashina A., Oyeniyi, Joshua O., Ojo, Adeolu O. ABSTRACT: Individuals have distinctive and unique traits that can be used to distinguish them from other human beings, acting as a form of identification. Human beings have always had the ability to recognize and distinguish between face features and with advent of machine learning, computers have been shown to have the same ability to recognize and distinguish between face features. In an attempt to develop a real-time facial recognition system, this study proposed a deep learning framework through the use of a pretrained convolutional neural network. The proposed framework was implemented on Matlab R2018a, using the Pretrained AlexNet convolutional neural network, using a locally sourced dataset; acquired through the use of the webcam. The average recognition accuracy of the system is 100%, while it takes 176 seconds to train the model and 2 second for verification KEYWORDS:Biometric, Face Recognition System, Deep Learning, AlexNet |
122 - 126
|
|
|
» Determinstic modeling to indentify the effect of HIV screening » Adeniyi Oyewole Ogunmola, Emmanuel Tejuola Jolayemi ABSTRACT: In this article, we present a deterministic modeling to identifying the effect of HIV screening. A varying proportion of the population is screened for HIV infection, and individuals with the infection are found and receiving treatment, the effect is measured on morbidity and mortality due to the disease in the population. The purpose of this work is to further show that screening a population for infectious disease like HIV is very essential to reduction or elimination of morbidity and mortality due to the disease. In this simulation work, HIV infected individuals at the asymptomatic stage are searched for by the screening before they would advance to the symptomatic stage. The results obtained showed clearly that when HIV screening is positioned at random sampling of the population in search of HIV individuals at the asymptomatic stage will much have better impact on morbidity and mortality due to the disease. KEYWORDS:Deterministic modeling, HIV/AIDS, HIV screening, asymptomatic stage, morbidity and mortality |
127 - 132
|
|
|
» Cubic transmuted lindley as a better distribution in the family of rayleigh distributions » Adesiyan Adefowope, Ifeyinwa Vivian Omekam, Adebowale Olusola Adejumo ABSTRACT: A new statistical model for non-normal data is proposed with some of its statistical properties as Cubic of the Transmuted Lindley distribution (CTLD), based on a new family of life time distribution. The new model contains some of lifetime distributions as special cases such as exponentiated Lindley, transmuted Lindley and Lindley distributions. These include the density and hazard rate functions with their behavior, moments, and moment generating function, skewness and kurtosis measures. Maximum likelihood estimation of the parameters and their estimated are derived. An application of the model to a real data set is presented and compared with the fit attained by other well-known existing distributions. KEYWORDS:Transmuted Lindley distribution; maximum likelihood method; transmutation map; hazard rate function, reliability function, parameter estimation |
133 - 142
|
|
|
» A simulation study on the performance of Bayesian and L2 regularization methods in
multicollinearity problem » Ojo O. Oluwadare ABSTRACT: The L2 regularization method known as ridge estimator had been widely used in literatures to solve the problem of multicollinearity. However, choosing an appropriate ridge parameter, k value makes this method to be difficult in application. This work carried out a simulation study to examine the performance of Bayesian, ridge and Ordinary Least Squares (OLS) estimators in the presence of multicollinearity to know the most efficient estimator. Mean Squared Error (MSE) was used as a criterion to judge the performance of these estimators. Results obtained from the simulation study revealed that Bayesian method is more efficient than the ridge and OLS estimators. KEYWORDS:Bayesian, Efficient, Multicollinearity, OLS, Regularization |
143 - 147
|
|
|
» Efficient new ratio estimators of population mean using Product of sample size and conventional location parameters » Sikiru Adetona Suleiman, Amos Adedayo Adewara ABSTRACT: This paper modified nine ratio-type estimators of population mean of the study variable from auxiliary variable suggested by Abid et al. [1] using product of sample size and conventional location parameters. The results obtained on the Mean Square Errors using the same data set used by Abid et al. shows that the newly modified ratio-type estimators perform better vis-à-vis the Abid et al. estimators, hence, recommended for usage in Sampling. KEYWORDS:Population, Estimator, Sample size, Parameter, Efficiency |
148 - 151
|
|
|
» Multivariate Analysis of Variance to Investigate the Effect of Nitrogen and Phosphorus on the Height, Weight and fruits Yield of Tomatoes » Osuolale Peter Popoola ABSTRACT: Balance fertilization in crops act as an insurance against possible nutrient deficiencies that may be created by the repeated use of single land for plantation. Many farmers apply different combination of chemical fertilizers like Nitrogen, Phosphorus and Potassium known as NPK fertilizers to enhance the height, weight and yield of tomatoes fruits. However, excessive use of fertilizers has an impact on the environment and human health. Despite the recognition of the negative effects associated with excessive fertilizer use, world consumption of fertilizers has continued to increase. This research work therefore, employed Multivariate Analysis of Variance (MANOVA) to investigate if nitrogen and phosphorus application contribute to the height, weight and yield of tomato fruit; identify specific level of nitrogen and phosphorus require for the optimal yield of tomato; and determine the optimal yield of tomato fruit at recommended levels. KEYWORDS:Multivariate Analysis of Variance (MANOVA), Analysis of Variance (ANOVA), Qualitatively Different Dependent Variables (QDDVs) and Tomatoes Yield |
152 - 156
|
|
|
» Image Processing Techniques based crowd size estimation » Mohanad Abdulhamid, Lwanga Wanjira ABSTRACT: Image processing algorithms are the basis for image computer analysis and machine Vision. Employing a theoretical foundation, image algebra, and powerful development tools, Visual C++, Visual Fortran, Visual Basic, and Visual Java, high-level and efficient computer vision techniques have been developed. This paper analyzes different image processing algorithms by classifying them in logical groups. In addition, specific methods are presented illustrating the application of such techniques to the real world images. In most cases more than one method is used. This allows a basis for comparison of different methods as advantageous features as well as negative characteristics of each technique is delineated. The main objective of this paper is to use image processing techniques to estimate the size of a crowd from a still photograph. The simulation results show that the different images have different efficiencies. KEYWORDS:Image processing, crowd size estimation |
157 - 163
|
|
|
» Prediction of the price of gold using autoregressive moving » Marius Constantin Popescu, Florentin-George Drig , Antoanela Naaji ABSTRACT: Nowadays, connectivity is an imperative factor within Intranet networks, especially in terms of speed, functionality, user-friendliness and time of intervention. The software application presented in this paper can be used to monitor network port(s), for one or several computers connected to the network, so that one could see, from a web administration page, which of the configured computers is in the network and which is not. This application can be an important tool for a small or medium business company because it simplifies the monitoring of the entire intranet network. The most important advantages are: high speed of updating the information, it does not slow the network by making useless traffic, it has a full reporting system and it is simple to use. KEYWORDS:network monitoring; user-friendly interface; connectivity; efficiency; flexibility |
164 - 172
|
|
|
» Automated Attendance Machine using Face Detection and Recognition System » Muhanned AL-Rawi ABSTRACT: This paper serves to automate the prevalent traditional tedious and time wasting methods of marking student attendance in classrooms. The use of automatic attendance through face detection and recognition increases the effectiveness of attendance monitoring and management. This method could also be extended for use in examination halls to curb cases of impersonation as the system will be able to single out the imposters who won’t have been captured during the enrollment process. Applications of face recognition are widely spreading in areas such as criminal identification, security systems, image and film processing. The system could also find applications in all authorized access facilities. KEYWORDS:Automated attendance machine, face detection and recognition |
173 - 182
|
|
|
» Some results on fourth-order integro-differential equations » K. O. Kareem, M. O. Olayiwola ABSTRACT: In this paper, we present the homotopy perturbation method to the fourth-order integro-differential equations. Some numerical examples are presented under the particular choices of the auxiliary parameters which can be used as a very accurate algorithm for solving linear fourth-order integro-differential equations problems. KEYWORDS:Homotopy perturbation method, boundary value problems, Taylor series, Integro-differential equations |
183 - 185
|
|
|
» An exploratory study of deep parsing approaches to automated text analysis and grammaticality evaluation » Moses Kehinde Aregbesola, Rafiu Adesina Ganiyu, Stephen Olatunde Olabiyisi, Elijah Olusayo Omidiora, Temitope Oyelakun ABSTRACT: Grammar-checking is a concept of interest in many natural language environments. It typically involves the evaluation of the grammaticality of transmitted messages, be it written or verbal. Automated grammaticality evaluation is the examination of natural language text for grammatical accuracy using computer software. The current study examined different deep parsing approaches adopted in computational linguistics for analysing the grammar of given texts. The study showed that the deep parsing approach proved to be quite precise in the outlining of grammatical rules. KEYWORDS:Computational Linguistics, Automated Text Analysis, Grammar Analysis, Grammaticality, Deep Parsing |
186 - 192
|
|
|
» Handling ties in survival time data with competing risk: a simulation study » Oladuti O.M, Abiodun A.A., Oyejola B.A. ABSTRACT: The Cox proportional hazards model is the most frequently method used to assess the effects of risk factors for survival time data when there are a single cause for the event occurring with the assumption that times are untied. Discrete-time logit model has been broadly used to model time-to-event data in order to handle ties but in a situation where more than two events are competing for the failure of an individual; only the first of them is observed, there is need to manage the ties. In this study, competing risk data using parametric model for cause-specific hazard and mixture of baseline hazard distribution were simulated, the performance of Cox model for the competing events and the overall were investigated with respect to different sample sizes in terms estimates of mean value, mean absolute bias (MAB) and mean square error for prediction (MSEp). The results showed that estimated mean values of covariate effects give estimates that converged to the true value of the parameter but Weibull-Exponential and Weibull-Gompertz yield a precise estimates with minimum MSEp compare to Exponential - Gompertz. The Efron approximation method gives an estimated means values that are close to the true values than Breslow approximation method. Discrete-time model performs better than Continuous time models in terms of estimated mean values with minimum mean square error as sample size increases. KEYWORDS:Cox PH model, Discrete-time logit model, Baseline hazard, Competing risks, Simulation Study |
193 - 198
|
» www.fcia.tibiscus.ro | » www.tibiscus.ro | | © 2003-2021 |