Pusat Pengajian Sains Matematik - Tesis

Browse

Now showing 1 - 5 of 462

A Novel Dynamic Evolutionary Model Integrating Discrete Hopfield Neural Networks With Satisfiability Problems And Its Applications In Image Encryption And Decryption
(2025-06)
Feng, Caicai
This thesis proposes a series of innovative DHNN-SAT variants and their applications. To address the inefficiency of traditional DHNN-SAT networks in solving SAT problems with dynamic constraints, a Dynamically Evolving Discrete Hopfield-SAT Neural Network with a flexible and scalable architecture is specifically designed. To tackle challenges posed by varying network scales and logical complexities, an optimized network based on a Crow Search Algorithm-guided Fuzzy Clustering Hybrid Method is proposed.
Improving Forecasting Accuracy For Time Series Data Using Fuzzy Techniques And Wavelet Transform
(2025-07)
Hobaykan A., Alenezy Abdullah
This study focuses on improving the accuracy of stock market forecasting for the Saudi Arabia stock exchange (Tadawul) by employing advanced modeling techniques and adaptive learning approaches. The study utilizes the maximum overlapping discrete wavelet transform (MODWT) in conjunction with various mathematical functions to analyze daily stock price indices data from October 2011 to December 2019.
Wh Factorizationanditsapplication In Graphenergies
(2025-01)
Mohamed, Bashir Dlal Juma
Hourglass matrix is a square matrix obtained from quadrant interlocking factorization, known as WH factorization, which can be represented in mixed graph. Recent establishment of hourglass matrix and its factorization technique does not show the all the properties of the matrix. In this thesis, properties of WH factorization were put forward which include its existence for every strict dominant diagonal matrix, that its HSYSTEM forms a lower triangular invertible matrix and its counterpart W-matrix forms an upper triangular invertible matrix. To optimize WH factorization, a variant cramer’s rule is used to show that it has an advantage over the traditional cramer’s rule. It can be concluded that the block WH factorization technique is able to partition the matrix for effective factorization. The matrix obtained from WH factorization is known to be well represented as mixed graph, but not as topological index graph. There are over 50 types of graph energies obtained from degree based topological indices, only eight types are considered for the mixed hourglass graph: first and second zagreb energy, first and second hyper-zagreb energy, first and second guorava energy, and first and second hyper-guorava energy. Furthermore, all considered energies are compared to conclude that the energies give value of even numbers while the graph energy ratio (ratio between laplacian energy and the mixed energy of mixed hourglass graph) is odd.
Isolated Signalised Intersection Optimisation Under Mixed Traffic Flow
(2024-11)
Zhu, Hongtao
Traffic congestion has become a severe problem. Numerous studies have focused on optimising intersection performance under homogeneous traffic flow, which refers to a situation on a road where vehicle dimensions do not vary much from each other and vehicles behave similarly in terms of speed and following distance. However, few studies have considered mixed traffic flow. Mixed traffic flow is exceedingly stochastic and non-linear, with numerous traffic compositions which limit the largescale application of previously proposed methods. Hence, the intersection performance under mixed traffic flow is optimised from two aspects: intersection layout and signal timing. For intersection layout, this study considers an existing one-lane left-turn bypass intersection. To improve the intersection performance, an alternative two-lane left-turn bypass intersection layout is proposed. A simulation study to compare the proposed intersection with the two other intersection layouts under mixed traffic flow is conducted. The results show that the proposed intersection outperforms the two other layouts. For signal timing, this study considers the multi-objective signal timing optimisation problem for isolated intersection under mixed traffic flow. The findings demonstrate that traffic signal plans yielded by the proposed method have more advantages in reducing traffic congestion compared to the existing traffic signal plans. A main contribution of this study is the consideration of mixed traffic flow by including two-wheeled vehicles in the model.
Pengautomasian Penganalisisan Nota-Nota Kaki Penyata Kewangan Berasaskan Pengetahuan
(1991-07)
Wan Abdul Razak, Rohaya Amal
Matlamat utama di sini adalah untuk memperolehi suatu rekabentuk model yang bersifat pedagogi bagi transformasil penjelmaan nota kaki penyata kewangan (dalam bentuk teks) ke dalam bentuk perwakilan pengetahuan berasaskan kerangka. Perwakilan implementasi yang dipilih untuk tugas ini ialah perwakilan Framelog [Zahran 87, 89, 90], yang pada dasarnya merupakan gabungan dua formalisma pengetahuan, iaitu struktur kerangka dan formalisma logik.

Browse

Recent Submissions