Mohamad Hanif Abd Hamid
Thumbnail Image
Preferred name
Mohamad Hanif Abd Hamid
Official Name
Mohamad Hanif, Abd Hamid
Alternative Name
Abd Hamid, Mohamad Hanif
Main Affiliation
Scopus Author ID
57197881764

Research Output

Filters

Reset filters

Settings
Now showing 1 - 2 of 2
  • Publication
    Design and development of multi-terrain mobile robot in large scale plantation
    ( 2015)
    Mohamad Hanif Abd Hamid
    The wide range applications of mobile robots can be seen from domestic appliances to large scale implementation. One of the possible applications that can benefit from the use of mobile robots is large scale plantations. However such applications, say in oil palm plantations, poses real challenge due to the multi-terrain nature of such environment. Described in this thesis is the development of multi-terrain mobile robot for oil palm plantation. The development of the robot consists of three different of prototypes which test the different design parameters of the mobile robots, and the analysis and results used for the design and development of AGROBOT. Several implementation strategies were tested, such as localizations, power consumption and ability to maneuver. The testing of the mobile robot was a success and able to move along desired preset paths along the oil palm trees. The waypoint navigation will follow the path and recorded the desire route with the capability of avoiding the obstacle. The success in the implementation of a multi-terrain mobile robot, AGROBOT, will benefit the agro industry and may be used for application such as pesticide spraying and weeding.
  • Publication
    Mitigating Overfitting in Extreme Learning Machine Classifier Through Dropout Regularization
    (Universiti Malaysia Perlis, 2024-02-14)
    Fateh Alrahman Kamal Qasem Alnagashi
    ;
    Norasmadi Abdul Rahim
    ;
    Shazmin Aniza Abdul Shukor
    ;
    Mohamad Hanif Abd Hamid
    Achieving optimal machine learning model performance is often hindered by the limited availability of diverse datasets, a challenge exacerbated by small sample sizes in real-world scenarios. In this study, we address this critical issue in classification tasks by integrating the Dropout technique into the Extreme Learning Machine (ELM) classifier. Our research underscores the effectiveness of Dropout-ELM in mitigating overfitting, especially when data is scarce, leading to enhanced generalization capabilities. Through extensive experiments on synthetic and real-world datasets, our findings consistently demonstrate that Dropout-ELM outperforms traditional ELM, yielding significant accuracy improvements ranging from 0.19% to 16.20%. By strategically implementing dropout during training, we promote the development of robust models that reduce reliance on specific features or neurons, resulting in increased adaptability and resilience across diverse datasets. Ultimately, Dropout-ELM emerges as a potent tool to counter overfitting and bolster the performance of ELM-based classifiers, particularly in scenarios with limited data. Its established efficacy positions it as a valuable asset for enhancing the reliability and generalization of machine learning models, providing a robust solution to the challenges posed by constrained training data.
      15  1