"Urban Green Space Access and Mental Health Impact in Hampton Roads" is an impactful project developed for the Hampton Roads Datathon hosted by the City of Norfolk. This initiative aimed to examine how urban green space availability correlates with mental health outcomes in Hampton Roads, utilizing data science to identify areas that could benefit most from enhanced green infrastructure. Leveraging Python, GIS-based spatial analysis, and publicly accessible datasets—including GIS data on parks and health statistics from sources like the CDC—the project mapped green space accessibility against mental health metrics across various demographic groups.

Through dynamic mapping and geospatial analysis, the project highlighted "green space deserts" in underserved communities where new parks or recreational facilities could positively impact mental well-being. Intended as a resource for city planners and public health officials, this study showcases the potential for data-driven insights to promote sustainable urban development and improve community mental health. The findings offer a compelling argument for targeted green space expansion, reinforcing Hampton Roads' commitment to a healthier, more resilient urban environment.

Stock Price Prediction and Portfolio Management is an innovative project developed as part of the Data Science course DS620 at Old Dominion University during the Spring 2024 semester. Crafted entirely in Python and leveraging advanced machine learning techniques, particularly Long Short-Term Memory (LSTM) networks, this project operates seamlessly as a backend system within the Google Colab environment. By analyzing historical stock price data derived in real-time from Yahoo Finance, the system employs LSTM models to forecast future stock prices, providing insightful recommendations on investment strategies. 

Despite lacking a traditional interface, Stock Price Prediction and Portfolio Management offer users a powerful tool to make informed investment decisions. Its integration with Google Colab showcases the adaptability and scalability of the solution, making it accessible and convenient for users to utilize. Rooted in academic rigor and driven by practical application, this project embodies the innovative spirit of data science, paving the way for advanced data-driven approaches to financial management. 

ODIS represents a straightforward yet robust Machine Learning-driven chatbot. It harnesses a live video feed to analyze an individual's emotional state and, based on real-time user feedback, embarks on a conversation by generating prompts. These prompts are dynamically created as the interaction progresses. After a few initial prompts, the chatbot autonomously generates prompts to enhance the user's comfort and engagement.

Subsequently, this concept materialized into a prototype, which underwent rigorous testing, resulting in an impressive accuracy rate of 74% and a remarkable success rate of 82%. As a testament to its innovative nature and remarkable success, ODIS was later submitted as the final year project and earned the distinction of being recognized as one of the most innovative projects in its category.

RCPC is meticulously designed to assist individuals with physical disabilities in easily accessing electrical devices, reducing their reliance on others. This innovative solution, RCPC, is a user-friendly, plug-and-play IoT module comprising two essential components: a transmitting remote and a receiver modem that seamlessly interfaces with the switchboard.

Notably, configuring the modem to the switchboard is a quick process, taking just 5 minutes to set up. With this minimal setup time, RCPC is ready to deliver enhanced convenience. This impactful project was later submitted as my mini-project during my Bachelor's Degree.

Multi-Hash, born from a visionary idea, was designed to elevate online account security amid concerns like phishing and rainbow tables. While industry leaders emphasized multifactor authentication and strict password rules, Multi-Hash offered a cost-free solution.

This algorithm operates in two stages: locally hashing the password on the user's device using a hashing algorithm, then adding a layer of protection by utilizing the inline sequential algorithm for salting and hashing it on the server before storing it in the database. This way the password is never exposed as raw data. In case of a breach, hacking becomes a formidable challenge due to Multi-Hash's approach to password handling.

This groundbreaking project demonstrated a remarkable 79% effectiveness. It was chosen to represent our college in the prestigious Smart India Hackathon and secured a top-10 spot in a state-level 24-hour hackathon. Multi-Hash exemplifies innovation and robust online security.