Data Analyst Python

Saketan Patil

Address: ***, ******* ******, ********, ** 07029

Phone: +1-551-***-**** Email: *********@*****.*** LinkedIn: https://www.linkedin.com/in/saketan/ Education:

New Jersey Institute of Technology, Newark, NJ Sept 2018-May 2020 Master of Science - Data Science GPA: 3.8/4

• Academic Courses: Machine Learning, Deep Learning, Data Structure and Algorithms, Introduction to Big Data, Database Management Systems, Applied Statistics, Data Analytics with R Programming, Data Mining. Manipal Institute of Technology, Manipal, India July 2013- July 2017 Bachelor of Technology - Computer and Communication Engineering GPA: 7.46/10

• Relevant Academic Courses: Statistical Analysis and Applications, Machine Learning, Data Mining and Predictive Analysis, Hadoop Programming Lab.

Skills:

Python (Pandas, Scikit-learn, NumPy, Keras, TensorFlow, Pytorch, opencv, scipy, nltk), R, Rcpp, Java, C#, .net, Windows, Linux, Unix, Tableau, Database Management with MySQL and Oracle DB, AWS EC2, Visual Studio, Rapid Miner, Hadoop Platform using MapReduce, HIVE, PIG, GPU Programming using CUDA, OPENMP. Experience:

GTBM/ Info-Cop East Rutherford, NJ Machine Learning, AI Intern Jan 2020- May 2020

• Explosive and Narcotic Prediction: Analyzed waveform parameters of chemical substances to detect hazardous and illegal materials with 85% accuracy. Design and optimization of Machine Learning models for Multi class classification.

• Crime Prediction: Implemented Exploratory Data Analysis of Crime data of 20 years. Application of Long Short Term Memory to identify patterns and crime count based on time series analysis. Algorithm implemented with mean absolute percentage error of 14%.

• Deployment of ML Models and applications in production environment (.net MVC). Launched Crime Count Predictor Web Application (.net MVC) which displays future crime count based on region. AKA, New York City, NY Data Analyst Intern Sep 2019- Dec 2019

• Worked on insights of Broadway data to analyze revenue, popularity and future of Broadway shows.

• Application of Natural Language Processing to recommend Broadway shows to users. Performed Sentimental analysis using Naïve Bayes algorithm to determine Show Category

• Predicting and comparing weekly statistics of a shows from Google Analytics to estimate growth or decrements in revenues.

• Dashboard development using Tableau to present Broadway show sales patterns to clients. New Jersey Institute of Technology, Newark, NJ Machine Learning Research Assistant Feb 2019- Aug 2019

• Evaluation of Deterministic and Non-Deterministic Clustering Algorithms on different toolkits and libraries.

• Analyzed the source code of algorithms and spotted merits and demerits in them with respect to ARI scores.

• Used self-organizing feature map I.e. Artificial Neural Network Algorithm for dimensionality reduction and clustering to demonstrate data patterns and accuracy behavior on 528 medical datasets. Wipro Technologies, Bangalore, India Project Engineer Jan 2017- July 2018

• Worked on Network Configuration and Quality Assurance. Evaluated the performance of network signals by parameters.

• Development of Ericsson 5G Indoor Radio System. Experience in Java, Python, LTE, MySQL and Linux OS.

• Full Stack Developer: Study of datasets and developing their representation in the form of dashboards. Publications:

• Saketan Patil, Xin Yin, Iulian Neamtiu, Sean Andrews. [2020]. Implementation-induced Inconsistency and Nondeterminism in Deterministic Clustering Algorithms. International Conference on Software Testing, Verification and Validation (ICST). Academic Projects:

• Application of Machine Learning on UCI Datasets using Bayesian Statistics, SVM, Logistic Regression, Perceptron, Multi-layer Neural Network, Decision Tree, Random Forest and Clustering Algorithms. Application of Feature Selection using Univariate

(Chi-square, F-score) and Multivariate Methods (PCA, LDA) on UCI Datasets to achieve higher accuracy.

• Development of Convolutional Neural Network from Scratch using optimizers like SGD, Adagrad, Adam. Usage of pre- defined models in Keras (Resnet150, InceptionV2) on image datasets like Ciphar, Imagenet, flowers, fruits for high accuracy.

• Query Sentiment Relationship Analysis: The project is a web application that finds the sentiment associated with the user query using a real-time twitter data. This is implemented on a Hadoop platform using Hive, MapReduce, SQL.

• Kaggle ASHRAE – Great Energy Predictor 3 Python: Built counter factual models across four energy types based on historic usage rates and observed weather. The data set includes three years of hourly meter readings from over one thousand buildings at several different sites around the world.

Contact this candidate