Abhinav Ram Bhatta Email & Phone Number

Chicago, Illinois, United States

Regression Analysis for The Department of Mental Health, Illinois• Directed the application of NLP techniques and regression models to analyze patient trends, achieving model accuracies of 81.32% (linear) and 89.77% (polynomial) regression.• Spearheaded the integration of supervised learning algorithms with large-scale medical datasets, reducing unnecessary hospitalizations by 13%.• Led the development of robust API frameworks to ensure seamless deployment of healthcare… Show more Regression Analysis for The Department of Mental Health, Illinois• Directed the application of NLP techniques and regression models to analyze patient trends, achieving model accuracies of 81.32% (linear) and 89.77% (polynomial) regression.• Spearheaded the integration of supervised learning algorithms with large-scale medical datasets, reducing unnecessary hospitalizations by 13%.• Led the development of robust API frameworks to ensure seamless deployment of healthcare analytics models, supporting real-time access for medical professionals.End-to-end Pipeline for Synthetic Syndromic Surveillance System on Azure Cloud platform• Architected and led the implementation of an end-to-end machine learning pipeline on Azure, processing over 2 TB of data to support real-time health surveillance.• Managed complex metadata for real-time syndromic surveillance using ArcGIS and vector databases, organizing geospatial data and health metrics to support faster data retrieval and enhance system responsiveness in crisis situations, reducing manual analysis time by 50%.Baseball Analysis for UIC Flames• Led the development of strategies using transformers (GPT-3) and NLP to optimize query rewriting, resulting in a 15% increase in predictive accuracy and a 35% improvement in operational efficiency.• Managed the end-to-end implementation of data pipelines using Python, scikit-learn, and Databricks, ensuring smooth and efficient handling of large datasets.• Utilized Databricks and SQL to design logical data models, managing over 3,000,000 records in a structured database to enhance metadata accuracy for efficient retrieval and data transparency for coaching staff and analysts. Show less

Hyderabad, Telangana, India

Analysis of Healthcare Reviews: Text Mining and Sentiment Analysis • Developed and implemented advanced NLP models and transformer-based techniques (GPT) to analyze healthcare reviews, achieving 89.56% accuracy in sentiment classification.• Led multiple projects focused on enhancing health threat detection across various medical domains, achieving a 20% improvement.• Automated critical reporting and data processing workflows using Databricks, reducing manual effort by 88%.•… Show more Analysis of Healthcare Reviews: Text Mining and Sentiment Analysis • Developed and implemented advanced NLP models and transformer-based techniques (GPT) to analyze healthcare reviews, achieving 89.56% accuracy in sentiment classification.• Led multiple projects focused on enhancing health threat detection across various medical domains, achieving a 20% improvement.• Automated critical reporting and data processing workflows using Databricks, reducing manual effort by 88%.• Applied PySpark and Databricks for data curation, integrating text mining models. Credit Analytics and Automation Enhancement • Enhanced and automated credit analytics models using advanced machine learning techniques, resulting in a 20% increase in model accuracy and a 66% overall accuracy rate.• Established data governance frameworks in Azure Cloud to ensure accuracy and reliability across automated analytics pipelines, reducing data processing times and enhancing the overall flow of data for strategic financial decision-making.• Utilized Docker for model deployment and CI/CD pipelines (GitHub Actions, Jenkins).Investment decisions for Individuals using Streamlit and Azure • Collaborated with financial analysts to define technical data requirements, leveraging GitHub Actions for CI/CD to deploy scalable solutions that supported real-time financial forecasting and streamlined data accuracy, contributing to a 20% improvement.• Utilized Docker for model deployment and CI/CD pipelines (GitHub Actions, Jenkins) for faster updates and continuous integration, reducing deployment time by 40% and ensuring a scalable solution.• Collaborated with financial analysts to integrate predictive modeling and supervised learning into the application, achieving a 93% user satisfaction rate. Show less