Ong Bi Lynn
Thumbnail Image
Preferred name
Ong Bi Lynn
Official Name
Ong, Bi Lynn
Alternative Name
Bi Lynn, Ong
Ong, B. L.
Ong, Ong Bi Lynn
Lynn, Ong Bi
Ong, Bi Lynn
Main Affiliation
Scopus Author ID
36728195700
Researcher ID
DKH-0788-2022

Research Output
Now showing 1 - 7 of 7
  • Publication
    ScanNote: a mobile application for enhanced text recognition and digital note-taking using machine learning-driven optical character recognition
    (Penerbit Akademia Baru, 2025)
    Wei Siang Hoh
    ;
    Nur Sofhia Izzaty Mohamad Azizi
    ;
    Ong Bi Lynn
    ;
    Si Kee Yoon
    ;
    Ma QiYuan
    ScanNote, a novel note-taking application designed to address limitations in current text recognition tools. Traditional OCR systems often struggle with accurately recognizing handwritten text, rotated images or text in noisy environments, creating inefficiencies for users needing reliable digital conversion of physical documents. ScanNote integrates machine learning (ML) with OCR to enhance accuracy and adaptability, offering a solution that outperforms traditional methods. The development of ScanNote responds to the demand for a more effective tool that can seamlessly convert both printed and handwritten text into editable digital notes. Evaluation results show that for printed text, ScanNote achieves 96.3% accuracy, compared to 86.7% for traditional OCR. When text is rotated 180°, ScanNote maintains 89.7% accuracy, while traditional OCR drops to 55.3%. For handwritten text, ScanNote reaches 84.1% accuracy, outperforming traditional OCR’s 54.5%. In addition to superior text recognition, ScanNote includes core note-taking functions and export capabilities, positioning it as a competitive tool in the digital note-taking market. Future research will focus on further improving accuracy for complex texts and optimizing real-time processing. ScanNote represents a significant step forward in bridging physical and digital note-taking.
  • Publication
    Use of natural language processing for the detection of hate speech on social media
    (Semarak Ilmu Publishing, 2025-09)
    Mehedi Hasan Shohan
    ;
    Kazi Rifat Ahmed
    ;
    Nur Farhan Kahar
    ;
    Nusrat Jahan
    ;
    Md. Maruf Hassan
    ;
    Nadira Islam
    ;
    R Badlishah Ahmad
    ;
    Ong Bi Lynn
    ;
    Naimah Yaakob
    Our society’s communication patterns have fundamentally changed as a consequence of the emergence of social media platforms. One effect of these changes is a rise in unpleasant behaviours like making rude and derogatory comments online. Speaking harshly or disrespectfully to someone in person may be difficult. However, online abuse and posting of improper material are considered to be acceptable. Hate speech has the potential to hurt a person or a group of people. Inappropriate material must be identified, in order to be filtered or banned from the web. CNN is a type of deep machine-learning model that has been suggested for such identification, because it performs better than conventional techniques in resolving text categorization problems. Our goal investigates how hate speech may be detected using NLP. In addition, a recent technique has been used in this field to a dataset. This classifier is assigned in each tweet to one of the three Twitter dataset categories of hatred, foul language, or neither. This model’s performance has been assessed with accuracy. The Naïve Bayes, the Decision Tree, KNN, Linear Regression, and the Random Forest are five algorithms that have been used. Of these, Linear Regression provided the greatest accuracy of 94%. It should be noted that when looking at each class separately, many hateful tweets have been mislabelled. It is advisable to look at the outcomes and faults in much detail, in order to comprehend the misclassification. Our analysis shows a better outcome in detecting hateful speech in social media.
  • Publication
    An Overview of Enhance Bandwidth Management in Cellular Network
    ( 2020-03-20)
    Isam H.M.
    ;
    Ong Bi Lynn
    ;
    R Badlishah Ahmad
    ;
    Elshaikh M.
    ;
    Al-Mejibli I.
    cellular networks have changed our life. However, current cellular network does not have an efficient bandwidth management during congestion. Some of these techniques was used have a few drawbacks such as the performance of the handoff mechanism depends on the channel quality, QoS, bandwidth, delay and speed of the cellular node, the handoff optimizing tend to increase the ping-pong handoff rate, frequency and bandwidth. The paper indicates the promising factors of the handoff mechanisms as decrease delay, enhance speed, saving energy and control bandwidth to achieve better throughput in wireless networks as compared to predecessor networks. How limited bandwidth are segregated among nodes is discussed in details in this paper. We believe that having implemented the proposed enhanced bandwidth management, congestion in current cellular network can perform better.
      1  20
  • Publication
    Consolidation of host-based mobility management protocols with wireless mesh network
    ( 2017-02-27)
    Wei Siang Hoh
    ;
    Ong Bi Lynn
    ;
    R Badlishah Ahmad
    ;
    Hasnah Ahmad
    The number of mobile devices increases exponentially and it becomes the trends and needs of human. With the increasing demands for new data and real-time services, there is a need of wireless Internet networks that can support different traffic characteristics and different Quality of Service (QoS) guarantees. All these Internet network mobile device services are supported by the mobility management protocols and its affects the QoS of the Internet. Having known the importance of mobility management protocol, we consolidate MIPv6, HMIPv6, FMIPv6 and FHMIPv6 with Wireless Mesh Network (WMN) into one environment. We identify, analyze, and compare the performance of Host-Based mobility management protocols integrate with WMN in terms of latency, throughput and packet loss ratio. At the end of this research, it is proven that the design and development of FHMIPv6 with WMN performs better as compared to the others Mobile Internet Protocols over the Internet using NS-2 Network Simulation software.
      3  24
  • Publication
    Performance evaluation of mobility management protocols in inter-network wireless mesh networks: a comparative study
    (Penerbit Akademia Baru, 2025)
    Wei Lok Wong
    ;
    Wei Siang Hoh
    ;
    Ong Bi Lynn
    ;
    Azamuddin Abdul Rahman
    ;
    Nor Syahidatul Nadiah Ismail
    This study addresses the critical challenge of mobility management in Wireless Mesh Networks (WMNs), focusing on the performance comparison of various Mobility Management Protocols (MMPs). The diverse range of MMPs, including Proxy Mobile IPv6 (PMIPv6), Fast PMIPv6 (FPMIPv6), Mobile IPv6 (MIPv6), Hierarchical MIPv6 (HMIPv6), Fast MIPv6 (FMIPv6), and Fast Hierarchical MIPv6 (FHMIPv6), present distinct advantages and limitations in terms of performance, scalability, security, and flexibility. Selecting the most suitable MMP for a given WMN application is a complex task, primarily due to the lack of comprehensive evaluation metrics that accurately reflect network efficiency. Furthermore, quantitative performance comparisons among these protocols are notably scarcen in existing research. Therefore, this research aims to contribute by categorizing MMPs applicable in Inter network WMN environments, developing a simulated WMN environment using the Network Simulator (NS-2), and conducting a thorough performance evaluation of various MMPs. Based on the result, the PMIPv6 outperform and able to provide 99.99% in packet delivery ratio, highest throughput compares to other MMPs. The outcomes of this study are expected to provide valuable insights for guiding the selection of MMPs based on specific application requirements, thereby contributing to the advancement of efficient and reliable WMN deployments
      1  2
  • Publication
    Internet of things (IOT) mobility support based on distributed sensor proxy MIPV6
    ( 2017-09-15)
    Jabiry M. Mohammed
    ;
    Ong Bi Lynn
    ;
    R Badlishah Ahmad
    ;
    Mohammed Hakawati
    It is expected that the footprint of the Internet of Things (IoT) will increase in the future. In order for this increase to occur, a network architecture which is flexible and capable of handling multiple flows with varying requirements as well as dynamically meeting the current demands. A new era of dynamic entities (nodes) within an environment like smart hospitals and cities is being empowered by IoT. The Sensor Proxy Mobile IPv6 (SPMIPv6) has been specifically determined for IP-based wireless sensor network (WSN) mobility with the aim of potentially reducing the consumption of energy by means of preserving the mobile nodes from being part of the handoff process. The majority of the shortcomings of the Proxy Mobile IPv6 (PMIPv6) such as non-optimized communication path, long handoff latency, and bottleneck issues were inherited by SPMIPV6. An improved SPMIPv6 architecture called Distributed SPMIPv6 (DSPMIPv6) is presented in this work with the aim of addressing the aforementioned problems. The solution proffered in the present architecture includes de-coupling the entities that are part of the control and data planes; the Dynamic Mobility Access Gateway (MAG) located close to the edge of the network distributes and manages the data plane, while the control plane, is dependent on a central entity called Sensor Local Mobility Anchor (SLMA). The introduced design is evaluated analytically, and the numerical results show that the performance of the DSPMIPv6 design is better than that of both SPMIPv6 and PMIPv6 protocols in terms of Local Mobility Anchor (LMA) load, and transmission cost performance metrics.
      23  2
  • Publication
    Performance comparison of energy efficient dynamic transmission and static transmission power in static mobility node wireless ad-hoc network
    ( 2017-12-11)
    Siti Asilah Yah
    ;
    Naimah Yaakob
    ;
    Ong Bi Lynn
    ;
    Mohammad Elshaikh Elobaid
    ;
    Wan Aida Nadia Wan Abdullah
    Transmission power optimization in Wireless Ad-Hoc Network is an important thing in order to minimize the energy consumption for effective utilization of the applications like Vehicle Ad-Hoc Network (VANET) applications. If one or more nodes in the wireless Ad-hoc network have little or no energy, then data transmission will be temporarily or permanently interrupted which might create a serious havoc in the Ad-hoc network especially during vital information transferred. This will, in turn, affect the performance of the entire network. Therefore transmission power control is one of the important research topics that needs to be focused in the wireless ad-hoc network in order to ensure effective energy consumption. Recently, we proposed a Dynamic Transmission Power algorithm to maintain network connectivity by adapting node's transmission power based on the distance between the vehicles in VANET. Our research aims to design a dynamic transmission power that can minimize the rate of energy consumption. Hence, in order to develop the proposed method, prerequisite experiment need to be done. This paper investigates the energy saving efficiency of dynamic and static transmission range in static mobility node wireless ad-hoc network which is prerequisite experiments before further experiment on VANET can be carried on. The simulation results prove that dynamic transmission range gives better energy consumption compared to static transmission range, so it is worth it to carry out the subsequent experiments on VANET.
      21  1