Portfolio

Awards

Total 24 Awards including:

• 3 Commissioner of Police Commendation and Testimonials in 2011, 2012 and 2018 for solving syndicated theft cases, excellent teamwork and best team in the Neighbourhood Police Centre and leading security operations successfully reducing crime cases by 76%
• 16 Commander Awards from 2012 – 2022 including led the arrest of wanted subject through teamwork and coordination, coordinating ground deployment exercises successfully, leading security operations leading to successful crime prevention in big jewellery event leading to zero theft case being reported etc
• 5 Director’s Awards from 2015 – 2017 for administrative efficiency and excellent teamwork

• One 9 Division Infantry Star Service Award in 2006 during National Service for service excellence, delivering quality service to customers during the course of work as an IT Support Specialist.

Aspect-Based Sentiment Analysis for Racial Issues in Singapore

Nanyang Technological University • 08/2022 – 11/2022

As the team leader in this pioneering project, I steered our efforts to analyse Singaporeans’ attitudes towards racial issues through aspect-based sentiment analysis on local social media, using a BERT-based deep learning model. We gathered and processed data from 2017 to 2022 from platforms like Hardware Zone and Reddit. This project aimed to understand public opinion on racial matters intricately, juxtaposing our findings with recent racial relations surveys to discern societal attitude trends and shifts. Despite challenges in crafting a balanced dataset from heavily negative web-scraped data, our analysis uncovered increasing negative sentiments in key areas such as employment, education, and housing. Our work shed light on public opinions regarding government policies and social issues, ultimately guiding strategies for societal improvement in Singapore.
Read more.

Using Machine Learning for Detecting Hacking Tools

Nanyang Technological University • 01/2022 – 05/2022

In this cybersecurity project, I utilised machine learning to detect hacking activities using forensic evidence from PCAP files, typically generated by packet sniffing programs like Wireshark. The project focused on translating these files into a human-readable format to identify suspicious activities such as DDoS and port-scanning, addressing the challenges of manual analysis due to the vast size and complexity of the data. We developed a supervised machine learning-based program that identifies common network intrusions like port scanning, operating system scanning, etc. This project demonstrated the potential of ML-based intrusion detection systems uncovering network anomalies with well-curated datasets that went through ETL and expert model optimisation.

Pair Trading Strategies with Machine Learning on S&P500 stocks

Nanyang Technological University • 08/2021 – 11/2021

In this AI project, I developed a machine learning program for pairs trading in the stock market, employing K-means clustering, Hierarchical Clustering, and Affinity Propagation to identify highly correlated stock pairs from the S&P 500. The key challenge was ensuring financial stability, which I addressed by integrating the Altman Z-Score formula to assess bankruptcy risk. This approach effectively narrowed down 28 potential trading pairs to 5 optimal ones, demonstrating my ability to innovate and harness complex ML techniques and financial analytics to minimise risk and optimise trading strategies in a dynamic market environment.