Business Data Analyst

Vijaya Sree Vignatha Vangala

***************@*****.*** 513-***-**** Cincinnati, Ohio LinkedIn Git Tableau Education

Master of Science in Information Technology, University of Cincinnati 08/2022 – 12/2023 Cincinnati, United States Machine Learning and Data Mining, Database, Advanced Storage Technologies, Cybersecurity Bachelor of Technology in Electronics and Communication Engineering, BVRIT HYDERABAD College of Engineering for Women

08/2018 – 07/2022 Hyderabad, India

Database Management Systems, Data Structures, Data Communications and Networking, Programming for Problem Solving, Scripting Languages Lab, Business Economics and Financial Accounting Professional Experience

MetLife Insurance, Data Analyst 08/2023 – Present Remote, United States

•Crafted, analyzed, and optimized complex SQL queries for policy and claims data, resulting in a 40% improvement in data retrieval efficiency and a 25% increase in query accuracy.

•Handled supervision of importing and exporting large actuarial data files (up to 5 TB a month) for use by underwriters in a range of file formats (CSV, JSON, Parquet) with 99.9% accuracy and consistency.

•Conducted requirements gathering and identified requirement gaps in insurance applications to ensure comprehensive project documentation and execution, leading to a 25% reduction in project delays.

•Utilized advanced SQL functionalities such as window functions, aggregate functions, joins, temporary tables, and views to extract meaningful insights.

•Collaborated with cross-functional teams to address data-related challenges, demonstrating flexibility and a growth mindset. Deloitte, Data Analyst Intern 09/2021 – 06/2022 Hyderabad, India

•Part of a data extraction migration initiative, moving from ‘SQL on Teradata’ to ‘PySpark on DataLake’ for data processing and visualization, resulting in ~50% effort reduction.

• Created SQL queries and automation programs with MS Excel’s Visual Basic Application for reporting on property occupancy and for QA, achieving a reduction in operational time of about 70 per cent and with no QA problems.

• Built reports by analysing large data sets with SQL and Python, pointing out major trends and patterns and raising success rate of data-informed decision making by 30 per cent.

• Performed extensive data analysis to drive decision-making about operational performance in response to our omnichannel promotion strategy.

• Created a weekly tracker with Tableau to assess store-level operation performance against omnichannel sales promotions for 10+ FBUs and physician-focused consumption over time.

•Utilized advanced SQL functionalities and Python for data manipulation to extract meaningful insights. Skills

Programming Languages

Python 3(Pandas, NumPy, Matplotlib, Seaborn, Scikit Learn, Beautiful Soup), SQL

Machine Learning

Covariance matrix optimization, Classification (Random Forest, KNN, SVM), Regression

Modeling (linear, sparse, logistic, regularized), Principal Component Analysis (PCA, PCR, sparse PCA), clustering

(K-means)

Technologies & Tools:

MS Excel, Google Sheets, MS PowerPoint, MS Access, Tableau, Power BI,DAX, Power Pivot

Stats & Experimentation

Time-Series Analysis (OLS, GMM, ARIMA, MLE), hypothesis testing, Monte-carlo

simulations, financial forecasting, Covariance, and correlation modeling

Projects

Medical Claim Denial Prediction 08/2023 – 12/2023

•Predicting reasons for rejection allows the companies to correct the denied claims and resubmit for a better approval rate.

•Successfully developed a machine learning model to predict medical claim denials with an accuracy of 85%

•Implemented a feedback loop system for continuous model improvement, ensuring the model adapted to new patterns and maintained high accuracy over time.

•Conducted training sessions for claims adjusters and underwriters, improving their ability to utilize model predictions effectively, thereby reducing manual processing time by 30%.

Email Spam Detection Using Machine Learning 01/2023 – 04/2023

•Developed a machine learning classifier to detect email spam with an accuracy of 95% and a precision of 92%,effectively reducing false positives and improving user experience.

•For spam identification, many machine learning methods such as Nave Bayes and K-Nearest Neighbor were applied.

•Conducted extensive hyperparameter tuning and feature engineering to optimize model performance, resulting in a robust and reliable spam detection system.

•Improved email security and user satisfaction by effectively filtering out spam emails, contributing to a cleaner and more secure inbox environment.

Contact this candidate