Research Papers - Dept of Information Technology

Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/593

Browse

Filters

2022 - 20262

Advanced Search

Filter by

Settings

Author: search.filters.author.Jayawardena, SStart date: 2022

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    PublicationOpen Access
    EuqAud: Detecting Gender Bias in Audio Datasets Using Polynomial Regression-Based Metric
    (Institute of Electrical and Electronics Engineers Inc., 2026) Jayawardena, S; Haddela, P.S; Shyamalee, T; Ekanayake, A; Mudalige, T; Dhanawardhana, I
    With the growing adoption of audio based AI systems in high-stakes domains such as healthcare, law enforcement, and social media, ensuring fairness particularly regarding gender bias has become critically important. While prior work on fairness has predominantly focused on disparities in model performance, bias inherent in training datasets remains underexplored. To address this gap, we propose EuqAud, a novel, pre-trained and traceable fairness metric that quantifies gender bias in audio datasets using raw acoustic features such as pitch, energy, amplitude, and voice activity. Unlike methods dependent on demographic labels such as race, age or language, EuqAud is designed to be demographic and language agnostic, enhancing its applicability across diverse contexts. The score is computed using an equation derived from polynomial regression with L2 regularization (Ridge regression), yielding robust and generalizable outputs. It spans a range from −10 to 10, where 0 denotes neutral, positive scores indicate male dominant bias, and negative scores reflect female dominant bias. For clarity, bias severity is categorized into three tiers: Neutral (EuqAud < 2), Moderate Bias (2 ≤ EuqAud ≤ 6), and Strong Bias (EuqAud > 6). Evaluation across multiple datasets demonstrates high predictive performance, with R2 values between 0.95 and 0.99. By focusing on dataset level bias rather than model outcomes, EuqAud offers a scalable and rigorous solution for advancing fairness in audio-based AI systems.
  • Thumbnail Image
    PublicationEmbargo
    ROS Supported Heterogeneous Multiple Robots Registration and Communication with User Instructions
    (IEEE, 2022-02-23) Rajapaksha, S; Jayawardena, S; MacDonald, B. A
    Different types of heterogeneous multiple service robots are working in the same environment to help humans in many ways in a smart house. These service robots have different capabilities based on the different control and communication systems. The complexity of the programming of the robots is now reduced by using middleware like Robot Operating System (ROS). However, the communication and control of the heterogeneous service robots using very high-level instruction is still tricky because of differences in ROS topics. If a user has issued a high-level instruction to all multiple heterogeneous robots in the same environment, then all robots must complete the given task without considering the software differences in each robot. This research has developed a web-based interface where users can input high-level instruction to all multiple heterogeneous robots running in the same environment. We have used three levels of instructions. Level 01 moves robot forward for the given speed for some distance without obstacles. Level 02 is to move the robot to a specific location without obstacles. Level 03 is to navigate the robot to a goal with some obstacles in the environment. Initially, all robots need to register their software specifications and hardware specification Format (URDF) in the robot registration engine with the help of the ontology. Then all service robots act according to the instruction given by the interpreter. The proposed system was evaluated using a simulated environment with a gazebo using "Turtlebot" and "Tiago" robots. Time complexity analysis of all algorithms was completed using the Big O notation.