Recent Submissions

Now showing 1 - 5 of 10
  • Item
    Multivariate Time Series with Application On Recurrent Neural Networks
    (Al-Quds University, 2020-06-12) Safa Nader Mustafa Shanaa; صفاء نادر مصطفى شناعة
    Multivariate time series data in practical applications, such as health care, geosciences, engineering, and biology. This thesis introduces a survey study of time series analysis to recurrent neural networks research, an analytic domain that has been essential for understanding and predicting the behavior of variables across many diverse fields, in this research the following were investigated. First, the characteristics and preliminaries of time series data are investigated and discussed, including various time series models, specially, Autoregressive Models such as, AR, MA, ARMA, and ARIMA. Frequently one wishes to fit a parametric model to time-series data and determine accurate values of the parameters and reliable estimates for the uncertainties in those parameters. It is important to gain a thorough understanding of the noise and develop appropriate methods for parameter estimation, so that various approaches of parameter estimates will be considered in this thesis, such as, yules walker method, least square method, method of moments and maximum likelihood approach. Second, different time series modeling techniques are surveyed that can address various topics of interest to artificial neural networks researchers, including describing the pattern of change in a variable, modeling seasonal effects, assessing the immediate and long-term impact of a salient event, and forecasting future values. The structure of the artificial neural networks especially the recurrent neural networks were discussed in details in this research, concerning on GRUs and LSTMs, and their properties, also some difficulties that arises in recurrent neural networks such as vanished gradient and the overfitting were discussed. To illustrate these approaches, an illustrative application based on Monte Carlo and bootstrapping methods is used throughout the research, constructing a one layer hidden recurrent neural networks and applied back-propagation, for the purpose of comparison, the variance of error in each method was estimated.
  • Item
    Linear Regression Models Assuming a Stable Distribution with Applications
    (Al-Quds University, 2020-08-31) Layla Khaled Mahmoud Lahaleeh; ليلى خالد محمود اللهاليه
  • Item
    Generalized Log-Logistic Proportional Hazard Model with Applications in Survival Analysis
    (Al-Quds University, 2022-08-10) Afnan Mahmoud Mustafa AL-sheikh; افنان محمود مصطفى الشيخ
    يمكن صياغة نماذج المخاطر النسبية Proportional Hazard مع أو بدون افتراض توزيع احتمالي لأوقات البقاء Survival Time. يؤدي الافتراض الأول إلى نماذج بارامترية Parametric، في حين أن الأخير يؤدي إلى نموذج كوكس Cox شبه البارامتي الذي يعد الأكثر شيوعًا في تحليل البقاء على قيد الحياة. ومع ذلك، قد يؤدي النموذج المعياري إلى تقديرات أكثر كفاءة من نموذج كوكس في ظل ظروف معينة. عدد قليل فقط من النماذج البارامترية في ظل افتراض PH تكون مغلقة، وأكثرها شيوعًا هو Weibull الذي يستوعب اقترانات الخطر أحادية الاتجاه فقط في هذا البحث سوف ندرس ونتوسع في تعميم التوزيع اللوجيستي الذي ينتمي إلى عائلة PH. لها خصائص مشابهة لتلك الخاصة باللوجستيات ، وتقترب من Weibull في النهاية. هذه الميزات تمكنه من التعامل مع اقترانات الخطر monotone and nonmonotone hazard function .تم تطبيق هذا النموذج على مجموعات مختلفة من البيانات وكذلك دراسة محاكاة، تبين أن النموذج يمكن أن يكون مفيداً جداً في وصف الأنواع المختلفة من بيانات الوقت إلى الحدث time-to-event بشكل مناسب.
  • Item
    Analysis of radiant heat exchange and interaction with conduction
    (Al-Quds University, 2009-04-09) Imad Ali Hamdan Al Atrash; عماد علي حمدان الأطرش
    This work deals with the three fundamental concepts of heat transfer modes (radiation, conduction and convection) with great emphasis on heat radiation and interaction with conduction. Determining radiation interchange between surface areas is needed in heat transfer, illumination engineering and applied optics. In fact, since 1960 the study of radiant interchange has been given impetus by technological advances that provides systems in which thermal radiation is very important. The geometric configuration factors derived here are an important component for analyzing radiation exchange. The computation of configuration factors involves integration, either analytically or numerically, over the solid angles by which surfaces can view each other. Some examples are given to demonstrate analytical integrations arise in radiant heat analysis. Moreover, the question of coupling heat radiation with conduction has been dealt with. To analyze this problem we consider a conductive - radiative heat transfer model containing two conducting and opaque materials which are in contact by radiation through a transparent medium bounded by diffuse - grey surfaces. Some properties of the radiative integral operator are presented. The question of existence and uniqueness of weak solutions for this problem is investigated. The existence of weak solution is proved by showing that our problem is pseudo-monotone and coercive. The uniqueness of solution is proved using some ideas from the analysis of nonlinear heat conduction.
  • Item
    Approximate Solutions of Einstein Field Equations
    (Al-Quds University, 2017-05-21) Mousa Saaed Mohammed Emter; موسى سعيد محمد مطير
    Einstein eld equations (EFEs) play an important role in understanding the theory of general relativity and related phenomena such as gravitational waves. Since, in general, it is almost impossible to nd analytical solutions of EFEs, it is necessary to solve these equations numerically (approximately). In this work, we derive the Einstein eld equations (EFEs) and the standard ADM (Arnowitt, Deser and Misner) equations form of EFEs. The ADM form consists of constraint equations and evolution equations for the raw spatial metric and extrinsic curvature tensors. The corner stone in the derivation of this form is \3+1 formalism", where one splits spacetime into three-dimensional space on the one hand, and time on the other. We study the BSSN (Baumgarte, Shapiro, Shibata and Nakamura) formulation. In this formulation the ADM equations were modi ed by factoring out the conformal factor and introducing three connections. The evolution equations can then be reduced to wave equations for the conformal metric components, which are coupled to evolution equations for the connection functions. Small amplitude gravitational waves were evolved and a direct comparison of the numerical performance of the modi ed ADM equations with the standard ADM equations was made. The results demonstrate that the standard implementation of the ADM system of equations, consisting of evolution equations for the bare metric and extrinsic curvature variables, is more susceptible to numerical instabilities than the modi ed form of the equations based on a conformal decomposition as suggested by Shibata and Nakamura. Further, in this work, we consider the problem of specifying Cauchy initial data in the case 3+1 formalism. We also apply the Optimal Homotopy Asymptotic Method, OHAM, and solving the Einstein eld equations corresponding to Schwarzschild geometry, i.e. we determine the Schwarzschild solution using OHAM.