DOMAINS
New domains are constantly shaping what is done today. I am excited by topics like computer vision, generative AI & LLMs. They will play a pivotal part in shaping the future & daily life.
New domains are constantly shaping what is done today. I am excited by topics like computer vision, generative AI & LLMs. They will play a pivotal part in shaping the future & daily life.
I love interacting with people about projects, concepts & research topics. These discussions have broadened my research interests & helped me improve & refine my understanding.
Attending lectures, forums & conferences has given me insights into trending topics, new domains & most importantly- new opportunities & connections.
The heart of Computer Science. Everything revolves around data. 99% of what I've learned & gained is by understanding & studying data.
Simulating different parameter values & permutations to compare random network models (Erdos-Renyi, Barabasi-Albert, Watts-Strogatz). Tested parameters include degree distribution, clustering coefficient & path length. Useful for selecting the appropriate network model for an Online Social Network (OSN).
Using machine learning algorithms to predict the likely disease & potentially affected organ based on the entered symptoms. NaΓ―ve Bayes, Random Forest & Decision Tree provide the best precision & accuracy. The application has been deployed as a web-app using Streamlit Share.
Data analysis of IPL match data since its inception. Used Python libraries to analyze data, & plot graphs & charts about various statistics including runs, wickets, records & net run rate (NRR).
Performed penetration testing to identify vulnerabilities associated with a WordPress Server. Identified solutions to the vulnerabilities (ex: Safeguarding user accounts & passwords).
Implementation of a database management system to store ticketing information for NBA matches. The front-end has been designed using NetBeans IDE. MySQL database serves as the backend. JDBC Driver facilitates the intermediate connection.
The application of reinforcement learning enables the Pacman agent & crawler to find the optimal path to their respective goal state. It can be executed by manual tryout or simulation of a fixed number of episodes.