Varun Muralidharan linkedin.com/in/varunmuralidharandata
512-***-**** addpxp@r.postjobfree.com Santa Clara, CA github.com/varunmuralidharan
EDUCATION
University of Texas at Dallas M.S. Business Analytics – Data Science (Honor’s Student) 3.70/4.00 May 2020
Course: Machine Learning, Business Intelligence, Predictive Modeling – Marketing Analytics, Big Data, Statistical analysis
Anna University, India B.E. Computer Science and Engineering April 2018
CORE COMPETENCY
Communication Problem Solver Team Worker Critical Thinking Active Learner Technologist Requirement Gatherer
TECHNICAL SKILLS
Programming: Python (scikit-learn, NumPy, Pandas, TensorFlow - Keras, SpaCy), R, H2O, SAS, SQL, Hadoop, Spark, Hive, git
Machine Learning: Regression, Clustering, Classification, Ensemble methods, decision trees, K-NN, SVM, Naïve Bayes, PCA
Databases & Analytics: Azure ML, MySQL, MS SQL Server, PostgreSQL, NoSQL - MongoDB, Tableau, Power BI, MS Excel, OBIEE
WORK EXPERIENCE
Essilor of America, Dallas, TX
Data Analyst - Consumer experience strategy & Directions May 2019 – November 2019
Achieved a 3.3% increase in target marketing by collaborating with the marketing team to segment customers using hclust
Developed business strategy by framing subscription plans to ensure 2X patient retention and effective treatment delivery
Identified potential $4.2 million markets and used two-stage least square to report price sensitivity for premium lenses
Performed ETL on 8 million health care data transactions, reporting sales, generate insights and suggested areas to improve
Designed and built data model for R&D to facilitate advanced querying from multiple databases
Ericsson, India
Analyst - Research & Development May 2018 – July 2018
Proposed plan to improve the lives of approximately 3.5 million legal refugees migrating into Turkey using call record data
Successfully increased clustering efficiency of regions by 4%, using well-being index data and weighted K-means method
Proved hypothesis on refugee behavior and distribution pattern by applying statistical tests such as t-test, ANOVA
Effectively created 18+ dashboards using Tableau, R, Google maps API to depict characteristics of refugees and cities
PROJECTS WORKS
Amazon Review Polarity Analysis (Python, text mining, Word2vec, TFIDF, NLP - NLTK, pipelining, VADER):
Improved the accuracy of review classification by 9% using Natural Language Processing (NLP) - VADER to find polarity, accommodating impact of expressions and semantics in reviews
Wendy’s Revenue Augmentation - Predictive Modeling (SAS, correlation, data visualization - Power BI):
Lead a team of 4 people, proposing a methodology to increase revenue up to 9% by identifying KPIs, consumer segments (RFM), price sensitivity, product combinations and expand loyalty base by optimizing promotional activities
Smart Transit System ((Published in IJPAM) R Shiny, JavaScript, MySQL, time series forecasting - ARIMA, GLM, dplyr, caret):
Created an application and a dynamic analysis dashboard to improve public transportation using R shiny and cloud server
Identified trends of commuters to optimize the frequency of routes, reporting live status, ETA along with seat availability
Vehicle Loan Default Prediction (Python, Scikit-learn, decision tree, AdaBoost):
Classified the customer’s ability to pay the installments on time, using credibility record and profile in the banking domain, and increased the accuracy by applying advanced algorithms such as bagging and boosting models
CERTIFICATIONS & COURSES
Neural Networks & Deep Learning - Coursera, DB5 - Stanford Lagunita, Adobe & Google Analytics, A/B testing, Jira - Agile