- (Present) Software Engineering, Undergraduate Degree, Singapore Institute of Technology
- Chemical Engineering, Diploma, Temasek Polytechnic
Lab Analyst Intern @ Syngenta Asia Pacific Pte Ltd
- Analysed pesticides using High-Performance Liquid Chromatography (HPLC) & Gas Chromatography (GC) for quantitative analysis
- Calibrated and validated equipment such as pH meter & Karl Fischer Titrators
Project Overview: A-Learn+ is an innovative sentiment analysis framework designed to extract meaningful insights from adult learning feedback by leveraging context-aware machine learning. This project focuses on utilizing advanced transformer-based models and custom neural network architectures to analyze feedback with a high level of contextual understanding, enabling precise sentiment predictions and actionable insights for enhancing educational experiences in adult learning environments.
- Transformer-Based Models: Fine-tuned transformer models like Twitter-RoBERTa and DistilBERT for context-based sentiment analysis, enabling the framework to capture subtle nuances in adult learning feedback.
- Custom Neural Architectures: Integrated advanced architectures such as attention-enhanced GRU, CNN, and FastText to complement transformer models by capturing contextual, spatial, and hierarchical sentiment patterns.
- Efficient Hyperparameter Tuning: Conducted sensitivity analysis on the number of trainable layers and learning rates to optimize model performance, balancing accuracy and computational efficiency.
- Data Preprocessing and Filtering: Scraped large-scale feedback data from social networks such as Twitter, Reddit and Tumblr, applied filtering mechanisms for relevance, and combined contextual post-comment structures to ensure robust training datasets.
A-Learn+ redefines sentiment analysis in niche domains by integrating context-aware machine-learning techniques tailored to adult learning feedback. This project empowers educational institutions and organizations to better understand learners' needs, refine their programs, and enhance learning experiences. By capturing context-specific sentiments, ContextLearn significantly improves the accuracy of feedback analysis, enabling actionable insights that drive informed decision-making and promote lifelong learning.
Project Overview: AiStudy is an innovative educational platform designed to enhance the learning experience through artificial intelligence and data-driven insights. This project focuses on leveraging advanced machine learning algorithms to personalize learning paths, thereby maximizing student engagement and educational outcomes.
- Machine Learning Integration: Developed and integrated machine learning models to analyze student performance data and adaptively suggest customized learning materials and quizzes.
- Data Analytics: Utilized big data technologies to process and analyze large datasets, allowing for real-time insights into student learning patterns and preferences.
- Interactive User Interface: Designed a user-friendly, interactive web interface using HTML, CSS, and JavaScript, which facilitates seamless user interaction and ensures a responsive learning environment.
AiStudy stands at the forefront of educational technology by providing a platform that not only responds to individual learning needs but also actively engages students in their educational journey. The adaptive learning technology implemented in AiStudy significantly improves learning efficiency, retention rates, and overall academic performance.
Project Overview: BookVerse is a cutting-edge online bookstore project that leverages both relational and non-relational databases to offer a personalized and efficient shopping experience. This platform is designed to handle a vast catalogue of books and efficiently manage user data to provide tailored recommendations.
- Database Integration: Utilized MySQL for managing relational data concerning book catalogues and customer information, and MongoDB for handling non-relational data like user preferences and recommendations.
- Recommendation Engine: Developed a sophisticated recommendation engine that analyzes user preferences stored in MongoDB to suggest books, enhancing user engagement and sales.
- Scalable Architecture: Designed and implemented a three-tier architecture, ensuring scalability and robust performance across presentation, application logic, and data storage layers.
BookVerse revolutionizes the way readers discover and purchase books by integrating advanced database management and analytics. This project not only improves operational efficiencies but also significantly enhances the user experience through personalized services.
Project Overview: SmartRoute is an innovative transportation solution designed to optimize the transfer experience from Singapore's Changi Airport Terminal 3 to various prestigious hotels. This system harnesses advanced algorithms to dynamically determine the most efficient routes, significantly enhancing fuel efficiency and reducing operational costs related to fuel and Electronic Road Pricing (ERP).
- Algorithm Development: Developed sophisticated routing algorithms that analyze real-time traffic data to optimize route selection, ensuring the most time-efficient and cost-effective travel.
- Business Intelligence Integration: Implemented a robust internal BI platform to support data-driven decisions in advertising and analytics, enhancing the overall strategy for route and resource management.
- Real-time Data Processing: Engineered solutions to process and execute real-time data models, facilitating immediate adjustments to routing based on current road conditions and ERP charges.
SmartRoute not only streamlines airport transfers but also serves as a model for integrating technological innovation into urban transport planning, demonstrating significant improvements in operational efficiency and passenger satisfaction.