Discover @abir-pattnaik Open Source projects

Abir Pattnaik (@abir-pattnaik)

abir-pattnaik

Stars
12
Global Rank 869,955 (Top 31 %)
Followers 6
Following 15
Registered almost 8 years ago
Most used languages

HTML
50.0 %

Jupyter Notebook
30.0 %

Python
10.0 %

R
10.0 %
Location 🇮🇳 India
Country Total Rank 36,757
Country Ranking

R
720

HTML
6,937

Titanic_Project_Udacity

This is a project done under the Data Analyst Nano Degree.

Retail-Fraud-Analytics

Hackathon Project for Retail Fraud Analytics

Udacity-EDA

This project was part of exploratory data analysis using R.We created functions that helped in univariate analysis and bivariate analysis.

Data-Analytics-Internship

This is a data analytics internship done as a remote internship under Sameer Mathur .This repository contains all the projects done during the internship.

Udacity-Inferential-Statistics

There are 2 type of statistics mainly -Descriptive and Inferential Statistics.This repository describes the Inferential Statistics that was done in this project.

Air_Quality_Analysis

This repository contains the air quality analysis of states of India.This contains EDA for the data set done.

IEDA

An interactive EDA that can help provide analysis from the data that was provided with you.I am building this project because in every ML problem a comprehensive exploratory data analysis is important thus I created a repository in which I know longer need to create the same figures over and again.

Jupyter Notebook

Piramal_Data_Science_Hiring

Model was created as part of Piramal Data Science Hiring Challenge.Target variables had 3 output variables with highly imbalanced dataset. Therefore, I had to use SMOTE to balance the dataset and build the model.

Jupyter Notebook

Suicides_in_india

This project was done as part of Global Hackathon conducted by Infosys in Bhubaneshwar.The data was collected from NCRB India.For us,this dataset was interesting for us because this is the first time we had used unsupervised learning and although it didn't yield any results it did help us develop our EDA.After a lot of brainstorming it was seen that the methods used in unsupervised learning were not satisfactory and thus we created probablistic model and showcase our results.Outputs can be found in this link -->https://www.kaggle.com/abirpattnaik/hackathon

Jupyter Notebook

Udacity-ML-Enron-Fraud

This is my first Machine Learning project and the learning curve rather grew a lot from here.I prefer to work in these kinds of project quite often that can enhance me.

Predictive-Underwriting

This project was done as part of Hackathon project conducted by Infosys.Underwriting (UW) is perhaps one of the most critical functions of Insurance Industry. Significant amount of cost and efforts are dedicated for this. With the advent of widespread use of Artificial Intelligence, there has been lot of interest in leveraging AI for UW. It has the potential to become a potent aid for Underwriters - less need of UW Resources per application, expedited UW processes, significant Cost and time savings.

Udacity-Data-Wrangling

This project was very important and one of the challenging parts as it involved cleaning of data and then applying queries on the dataset.