Pusat Pengajian Sains Matematik - Tesis

Browse

Recent Submissions

Now showing 1 - 5 of 452
  • Publication
    Curve Fitting With Bootstrap Error Analysis And Its Application On Two-Dimensional Data
    (2019-08)
    Sahubar Ali, Nur Soffiah
    In data fitting, researchers use various methods to determine the quality of a fitting. Visualization of images is crucial in observing the behavior of data obtained. The problem in judging the accuracy of a result obtained through visual observation are commonly faced by researchers when handling contaminated data such as noisy data, missing data and outliers. In this research, study has been conducted to deal with those noisy data and missing data using least square titting (LSF).
  • Publication
    Two-Dimensional Mathematical Model Named Tuna-Wq For Water Quality Analysis
    (2018-10)
    Chong, Michael Sueng Lock
    The main purpose of this study is to develop an in-house two-dimensional water quality model, codenamed TUNA-WQ, as an alternative to currently available water quality models, which are often either difficult to set up and limited in their application or are licensed.
  • Publication
    The Impact Of Gender And Gender Composition On Occupational Wage Inequality In Palestine: A Multilevel Modeling Approach
    (2021-03)
    Ayyash, Mohsen H.H.
    Inequality in the wage paid between males and females is a global phenomenon faced by the world. The problem is more severe in developing countries due to the social-cultural factor. Palestine is one of the countries experiencing a high ratio of imbalance participation rate of genders in the labor market in which female participation is among the lowest worldwide. On the other hand, occupational discrimination and wage inequality still exist between males and females. Combining both issues, this thesis seeks to examine the impact of gender and occupational gender composition on wage levels in the Palestinian labor market and contributes in three ways: first, this is the first study that analyzes the occupational gender wage gap in Palestine using multilevel linear models due to its high ratio of the gender pay gap. It does not limit the investigation of the betweenoccupation groups and within-occupation groups variations but also includes the examination between-gender-within occupation groups wage differentials. Second, this thesis demonstrates that the Bayesian estimator provides the most accurate and efficient estimation as compared to the conventional technique of maximum likelihood (ML) and restricted maximum likelihood (RML). Third, the thesis examines the effect of occupational sex segregation by utilizing the two-level wage models where occupations classification is decomposed using an ISCO-08 twodigit classification.This approach provides a more accurate estimate but not yet well-explored. The data are collected from the Palestinian Labour Force Survey (PLFS) over the period 2014 to 2018. The results reported evidence of wage inequality due to occupational groups which account for about 23.4% of wage differentials. On average, wages in male-dominated occupations are higher than those in gender-integrated and female-dominated occupations, which supports the devaluation hypothesis. The results also indicate that men enjoyed wage advantage over women across the gender-typed occupations, which supports the universal male advantage hypothesis. Moreover, the size of the gender pay gap is wider in occupations dominated by females as compared to other gender-typed occupations. Besides gender-based occupational discrimination, a significant portion of the between-occupation wage gap in the Palestinian labor market is mainly explained by workers’ characteristics including place of work, industrial sectors, sector of employment, region, work status, and marital status. The thesis suggests reducing the gender wage gap through equal pay enforcement and programs to encourage women's participation in the labor market.
  • Publication
    Panel And Spatial Panel Model Of Unemployment Rate For Selected Asian Countries
    (2024-07)
    Lim, Bao Man
    The unemployment rate has always been a concern in the global economic context. The main objective of this research is to examine the relationship between the economic freedom index (efi), index of governance quality, and economic indicators on the unemployment rate for selected asian countries. Therefore, this study explains and employs static panel data models, dynamic panel data models, and spatial panel data models to analyse the unemployment rate in the selected asian countries. The fixed effects model (fem) with robust panel and clustering is found to be the most suitable model for overall unemployment rate, male and female, which can address heteroskedasticity and serial correlation issues. Several of the economic freedom and governance quality indices were found to be significant in this study. Furthermore, spatial dependence is a condition where the outcome in one country depends on the outcome or other factors of another country. All test results indicate that the fem sdm for both effects (time and individual) for overall unemployment rate, male, and female is the most suitable model. The key factors that significantly affect the pattern of total and male unemployment rate of selected asian countries’ models, including government integrity, tax burden, monetary freedom, political stability no violence, and regulatory quality. While for the female unemployment rate, government integrity,
  • Publication
    Developing Hopfield Neural Networks Using Gaussian Distributed Small World Topology For Visual Object Tracking
    (2024-06)
    Sun, Jun
    Visual object tracking (vot) is considered a challenging research topic in artificial intelligence. Today, many industries rely on object tracking technologies to identify errors, monitor environments, and make timely decisions based on tracking results. Visual object tracking has enabled many innovations, such as autonomous vehicles, traffic monitoring systems, remote medical diagnostic systems, and more cutting-edge applications are on the horizon. However, among these notable achievements, it is worth noting that, unlike these object-tracking techniques, a human brain is more efficient for object tracking tasks and requires fewer resources. Recent neuroscience studies have shown that artificial neural networks organized as real cortical connectivity may perform more efficiently in complex recognition tasks. Therefore, a novel visual object tracking method based on hopfield neural networks is proposed in this study. A small-world network is employed as the topology of the neural network model. However, a biological feature is integrated into the small-world network model: the exponential decay rule, which may mimic some characteristics of the structure of the cerebral cortex. In the neural network, each pixel of video frames is assigned to a neuron at the corresponding position. Pixel strength is characterized as the state of a neuron. The video frame is memorized after all neurons in the neural network have been trained to a stable state. A bionic mechanism utilizing the associative memory property of a bionic hopfield neural network is proposed to track objects in video frames.