Data Analyst

Komal Sharma

• *** Main Street, ***** • Hartford, Connecticut • 860-***-**** • *****.******@*****.*** • LinkedIn Analytics professional with 3.5 years of experience in Data Science, generating insights, storytelling, automating, developing and compiling reports and dashboards using Tableau. Proficient in Statistics, ML, Statistical Modeling, R, Python, SQL, Tableau. EDUCATION

University of Connecticut - M.S. in Business Analytics and Project Management (GPA 3.78/4.0) Aug 2018 - Dec 2019 Maharishi Dayanand University - Bachelor of Technology in Computer Science (GPA 7.0/10.0) Aug 2010 - May 2014 TECHNICAL SKILLS

Machine Learning: Logistic / Linear Regression, Decision Tree, Random Forest, Boosting, KNN, Naïve Bayes, SVM, PCA, Clustering, Time Series models, Neural Networks, NLP, Validation techniques, Market Basket Analysis, Variable Transformation Statistical Analysis: Hypothesis Testing, A/B Testing, Data Cleaning, Data Mining, ANOVA, Quantitative Analysis Data Analysis: Exploratory Data Analysis, Feature Engineering and selection, Text Mining, Customer Analytics Organizational Skills: Project Management, Leadership, Team Building & Collaboration, Risk & Cost Management Tools: Tableau, R, Python, SQL, Alteryx, Linux, MS Excel, SAS Enterprise Miner, MS PowerPoint, MS Word, MS Visio, SAS JMP, Pentaho, SSMS, Hadoop, Big Data, TFS, Hive, Pig, Spark, Amazon Web Services[AWS – S3, EC2, Redshift), Google Analytics PROFESSIONAL EXPERIENCE

METLIFE ANALYST INTERN USA Jun 2019 - Aug 2019 Data Wrangling: Analyzed large dataset of ~44M claims in R (dplyr, tidyr, ggplot, caret) to provide trend analysis on medical procedures/ treatments administered and associated costs for 1st party claimants injured in auto accidents Developed analytical process for data loading, and identified key performance indicators like top providers and procedures by cost across countrywide/state

Data Modeling: Designed Logistic regression model for customer fraud claim detection with 78% accuracy Business Impact: This is fed into a chain of ‘fraud detection models’ for data driven decisions on better medical management strategies and preferred deductions with medical providers COGNIZANT TECHNOLOGY SOLUTIONS DATA ANALYST INDIA Aug 2015 – May 2018 People Analytics - Data Warehousing and Reporting

Work directly with stakeholders to perform business analysis and translated business needs into technology solutions Data Engineering Aggregated new hires data from multiple sources using ETL and SQL for data extraction and data manipulation. Loaded data into data warehouse reducing data access time for five internal teams Business Intelligence Created interactive Tableau dashboards for tracking KPIs of hired resources such as training cost per resource and quality of hire. Provided higher management a one-stop dashboard to track over 29 metrics Business Impact Provided Business Insights and recommendations to top management and cross- functional teams. Identified business metrics for efficient demand-supply management of resources and advised marketing strategies to senior management to reduce hiring cost by ~10% Trained and mentored new peers in multi-variate analysis, descriptive analytics and Entity- Relationship diagrams describing business model and helping them ramp up in technical and domain-specific business knowhow Worked on projects following Systems Development Life Cycle(SDLC) using Agile–Scrum methodology Predictive Modelling and Claim Analytics

Performed advanced predictive analytics by building statistical models like Logistic regression, Decision Trees and boosting using Python to find which of the expense claims of employees can be auto approved with accuracy of 88% Data Visualization: Developed Power BI dashboard and Excel pivots showing claims that comply with organizational rules for auto approval through classification models to reduce workload of auditors, thereby saving 10 man-hours TATA CONSULTANCY SERVICES ANALYST INTERN Apr 2014 - Jun 2014 Collaborated with data engineers to implement ETL process, wrote and optimized SQL queries to perform data extraction and merging from Oracle

Ad-hoc Analysis: Leveraged Excel to create Monthly Operational Reports (MORs) using pivot tables and Macros ACADEMIC PROJECTS

Customer Subscription model for Banking firm Machine Learning (Python) Analyzed bank customers to predict term deposit subscription with an accuracy of 83.69%. Minimized revenue losses occurring from false negative prediction to as low as 3.5%. Performed Data Cleansing, Correlation & Outlier Analysis, feature transformation in data preprocessing Fraud Claims Detection for Travelers Fraud Analytics (Python) Identified first-party physical damage fraudulence explaining reasons of fraudulent claims based on historical data. GBM model with 10% misclassification rate