Pusat Pengajian Sains Matematik - Tesis

Browse

Browsing Pusat Pengajian Sains Matematik - Tesis by Subject "Algorithms"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Publication
    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.
  • Thumbnail Image
    Publication
    Cryptocurrency Quantitative Trading Strategy Based On Machine Learning Approach
    (2025-06)
    Fu, Dingyu
    This thesis adopts the LSTM model to predict the price trends of cryptocurrencies by combining their historical prices with various technical indicators as features for research. We designed six control experiments to compare the impact of different technical indicators as features on the model. The final results indicate that the LSTM model combined with technical indicators can effectively improve prediction accuracy, but not all technical indicators contribute to the improvement of the model.
  • Thumbnail Image
    Publication
    Negative Based Higher Order Systematic Satisfiability Logic With Hybrid Black Hole Algorithm In Enhancing Multiunit Discrete Hopfield Neural Network
    (2025-06)
    Rusdi, Nur ‘Afifah
    Understanding intelligence is crucial for developing advanced intelligent models. In pursuit of this goal, satisfiability logical representation in Discrete Hopfield Neural Network has provided new insight in understanding the behaviour of the data. However, the role of negation in understanding intelligence has been overlooked as negation is often associated with false outcome. Negative Based Higher Order Systematic Satisfiability Logic is proposed to promote the appearance of negative literals within the clauses. The proposed logic demonstrates optimal performance as compared to existing logical rules. To further improve the quality of the final neuron states, Hybrid Black Hole Algorithm is proposed to update the neuron states that satisfy the multi-objective functions. The newly proposed mechanism will be incorporated into the logic mining model known as Multi-unit Negative Based Higher Order Systematic Satisfiability Reverse Analysis.