About Me
Mandhyan
I am a Graduate Student from Rutgers University - New Brunswick, major in Data Science. I am passionate and dilligent in upskilling myself and working on Machine learning, Deep learning & Data Science Technologies
I am a Graduate Student from Rutgers University - New Brunswick, major in Data Science. I am passionate and dilligent in upskilling myself and working on Machine learning, Deep learning & Data Science Technologies
Graduated from Rutgers University with a Master's Degree in Computer Science - Data Science, with 3.75 GPA (out of 4). With skillset including Deep Learning, Machine Learning, Reinforcement Learning, Natural Language Processing, Computer Vision, Data Visualization and Predictive Modelling.
September 2018 - May 2020Worked on building Anomaly Detection System to detect anomalous change in CMS NPI dataset.
February 2020 - May 2020Worked on building GPU based modules for Adaptive Real Time Machine Learning Platform (ARTML). Also optimized CPU based modules for faster performance.
May 2019 - September 2019Worked on Machine Learning projects including - Time Series Analytics and Computer Vision. Also deployed and administered Openstack Cloud Platform on Ubuntu Servers.
August 2016 - May 2018Graduated from SRM University with a Bachelor's of Technology in Computer Science and Engineering with 8.0 CGPA (out of 10). With skills in general programming, data structures and fundamentals of Computer Science. Also built an Eye Blink Counter Application on Andriod platform as my Final Year Project, this application was built to detect and keep track of human eye blink using Mobile front camera.
August 2012 - May 2016
As a Data Science Master's student at Rutger's University. I took upon an Independent Study under the guidance of Professor Charles W. Cowan (a great mentor). The goal of this study was to design and develop a Stock Trading Agent using Reinforcement Learning techniques which earn's reward/profit at the end of a time interval.
Covid-19 The first dataset which comes into one’s mind in today’s time, is very important but we are already looking at its analysis on the news like 10 times a day. Which made me think in the opposite direction and seek datasets which can be challenging as well as fun at the same time to display my skills gained in R. It reminded me of last year’s spring break, me and my friends took a road trip South. Towards Virginia, North Carolina to enjoy the best nature had to offer i.e. “The Smoky Mountains”. And taste different varieties of beer we could come across. During this trip we would stop to seek out famous local breweries and try different varieties of beer they had to offer. The process would always be to seek out breweries, in the area we were in, via the internet or through recommendation by locals; and learn about their types of beers, ratings and so on before visiting the place out. Now thinking back to those times I wanted to take a data Scientist approach to the same questions and seek answers using beer data sets.
As we all know, one of the biggest evolution of 21st Century is Social Media. It has gradually infiltrated our daily lifestyles. Be it Facebook, Twitter, Instagram, etc. People will find their friends, family or a complete strangers to connect and share their experiences with. This has given people the freedom, to express themselves to much wider audience than ever before in the history of Mankind. Social Media has bestowed us with a platform where people from different cultural backgrounds can socialize and be the part of Globally Acknowledged Citizen Movement. Who’s doing what and what should we do to be the part of the trend is the question which is chasing us for centuries? We answer these question by predicting trending topics on one of the biggest social network “Twitter”.
Collection of assignments done in Class STAT:581 Probability & Statistics for Data Science at Rutgers. Assignments - 1) Fit distribution on Category 4 & 5 Atlantic Hurricane Data, 2) Method of Moments Estimator, 3) Bootstrap & Jacknife Simulation, 4) Maximum Likelihood Estimator, 5) Bayes estimates & plots of the posterior distribution, 6)Final Project - To find the cancers which either over express or under express in the cancer and do further analysis, using NCI60 dataset.
People are looking for recommendations all the time. Be it, which movie to watch next or which book to read or which song to listen and so on. In todays world, technology has given us recommendation systems which provides users with personalized recommendation such as Netflix for Movies, Spotify for music, etc. So I learnt to build one of my own, this recommendation system recommends Anime to the user, its based on the Kaggle Competition data set and uses User-User based Collaborative filtering to recommend Ainme’s.
