One of 11 students selected for 2024-2025 academic school year.I will continue working with Dr. Madria and Ademola Adesokan in the Department of Computer Science. My research for this academic school year will focus on annotating political tweets, and building a model that can predict the voting preferences of a tweet using sentiment analysis.

• Web scraped and annotated over 27,000 disaster-related tweets with metadata (location, time, severity), applying text normalization and handling imbalanced data to build a high-quality dataset for first responder classification.• Developed and fine-tuned a RoBERTa-based machine learning model that achieved an accuracy of up to 98% in classifying first responders across various crisis events.• Built an interactive web application to visualize datasets and model predictions.

• Developed a comprehensive simulation program to model complex life cycles and behaviors of microorganisms in a dynamic environment, in collaboration with a team of biologists, ensuring biological accuracy.• Designed and built a user-friendly web application capable of running and comparing multiple simulationsconcurrently, with integrated real-time data collection and visualization, facilitating cross-disciplinary research.• Integrated an automated data pipeline with the web application for real-time data processing, storage, and analysis, improving efficiency and enabling deeper insights across biological models.

During my NSF-REU internship under the mentorship of Dr. Sid Nadendla, I contributed to a research project aimed at enhancing soldier performance by understanding cognitive workload and physiological responses. Our findings have significant real-world applications, such as developing advanced soldier assistance systems that monitor and respond to cognitive overload in real-time. The methodologies and technologies we developed also have broader applications in domains like healthcare, first responders, and other high-stress occupations.In this role, I developed a testbed using Python and MATLAB to assess cognitive load in real-time, integrating EEG (Enobio 20) and PPG (Polar Verity Sense) sensors. I ensured precise synchronization of physiological data collection with cognitive task performance through meticulous MATLAB scripting. Additionally, I led the front-end and back-end development for cognitive tasks such as the Flanker N-back and Multiple Object Tracking (MOT) tasks, using Python and JavaScript. My responsibilities also included developing efficient data processing and analysis pipelines to handle incoming EEG and PPG data in real-time.Looking ahead, I’m excited to work on the Aerial Drone Navigation task using the RealFlight Evolution simulator, followed by conducting advanced signal processing to analyze sensor data and developing a Convolutional Spiking Neural Network. This model will leverage EEG/PPG data and task performance metrics to predict cognitive workload.This project provided me with hands-on experience in integrating wearable sensors with web applications, real-time data synchronization, and advanced data processing techniques. These experiences have equipped me with the skills to handle complex, interdisciplinary projects and contribute to the development of innovative solutions in high-stress environmentsThe paper(s) developed from this research will be attached once they are completed and published.

• Developed responsive web pages using HTML, CSS, and JavaScript, ensuring mobile responsiveness, optimized load times, and adherence to web accessibility standards.• Collaborated closely with the design team to create visually appealing and user-friendly interfaces, resulting in a 20% increase in user engagement.• Integrated data analytics features to track user behavior, enabling data-driven decisions and real-time designadjustments that enhanced the overall user experience.