Name: Akarsh Singh
Type: User
Company: @Northeastern University
Bio: Data Scientist, Grad Student, Avid Researcher in the domains of ML, Deep Learning, and Stats. In a nutshell, I enjoy transforming data into valuable knowledge!
Location: Boston, MA
Akarsh Singh's Projects
This is a creative description of myself and the skills I possess. The main purpose of this repository is to let users know the kind of tools I've used and the projects I've done. It also showcases basic web-development skills and my interest in software technology
. This paper gives a brief information about the sorting of objects according to their color using TCS3200 color sensor, Arduino UNO and servo motors. The identification of color is done using frequency scaling of color detection.
SF Brigade's Data Science Working Group.
This project is a sub-part of a research publication co-authored by me where I have created models to predict cricket player performances and aggregate results to create the best fantasy team using multi-objective optimization techniques
This is a dataset created from webscrping, using python. Data was collected, cleaned, transformed, and aggregated from two websites () from over 20 tables. Beautifulsoup library in Python was used to achieve the same. Various predictive models were used to create an accurate predictor system. Players were chosen from top Football/Soccer leauges like Premier League, La Liga, Bundesliga, Serrie A, and Ligue 1. The final dataset had over 350 players with detailed statistics parameters of performance collected from over 2 seasons.
The purpose of this project is create a predictive model capable of identifying the probability of a person securing an admit based on their personal profile parameters. Simplified visualisations have been created for understanding the data. 80% accuracy was achieved on the test set.
This Notebook is based off an open source dataset available on www.kaggle.com where I have created models to classify patients who can potentially witness heart failure on the basis of various parameters. The best model had an accuracy of 94% and a recall of 91%
1st place solution for the Kaggle PANDA Challenge
Attempted production grade codebase for scalable and reproducible DL
Fully Functional OLTP Bank Management System
The purpose of this project is create a classification model capable of accurately predicting the price of secondhand cars. The data used for model building is open source and has been added to this repository. Most packages used are usually pre-installed in most developed environments and tools like collab, jupyter, etc. This can be useful for people looking to enhance the way the code their predicitve models and efficient ways to deal with tabular data!
Open Source Dataset used to Analyze Sentiments of Clothing Reviews Given By Customers.
This was a capstone project from a course I completed "Web Development - From Wireframes to Prototypes). Used tools like Illustrator and Balsamiq to create and brand a concept, luxurious coffee store called Doppio Experience. Implemented standard practices and standard practices to make the most of my learnings from this course.