Hymavathi Gummudala

Data Analyst

• Collaborate on workforce development projects by analyzing program outcomes and performance metrics to meet institutional goals.
• Design and maintain dashboards using Power BI to visualize data insights, ensuring data integrity and compliance with FERPA regulations.

Data Science Intern

Link to Repository:

• Extracted text and images from PDFs and DOC files, followed by semantic chunking using NLTK to create meaningful data segments.
• Generated embeddings for document chunks using a sentence transformer model to enable semantic understanding.
• Performed cosine similarity analysis between chunks, laying the groundwork for a Retrieval-Augmented Generation (RAG) pipeline.

Statistical Analyst (Lab Assistant)

• Facilitated student comprehension of statistical methodologies, such as ANOVA, t-tests, chi-squared tests,classification and regression analysis, through hands-on instruction and coding exercises.
• Course: INFO-I 415

Research Analyst

Causal Analysis Project

Link to Repository:

• Collected and preprocessed data from diverse sources (i.e, Huggingface, Github etc) , converting unstructured Text, JSON to CSV.
• Executed a 70-30 data split, facilitating comprehensive analysis across 110,000 sentences.
• Achieved an 81% accuracy with the advanced sequential algorithm
• Conducted detailed analysis, identifying instances of prediction flips and critical words.
• Executed parts of speech tagging using Spacy to analyze and determine the percentage distribution of each grammatical category
• Researched and completed a literature survey on mood changes regarding weather, Data Integration, Bio-medical Literature.

Text Extraction Project

Link to Repository:

• Spearheaded extraction and analysis of legal Latin phrases and case data from Supreme Court of Indiana documents, utilizing Python and Spacy techniques.
• Orchestrated comprehensive data preprocessing, converting PDF documents to text format and meticulously cleansing irrelevant information for streamlined analysis
• Executed exploratory data analysis (EDA) to uncover key insights, visualizing trends and patterns to inform strategic decision-making processes.
• Produced detailed reports and visualizations, effectively communicating findings to stakeholders and guiding informed business strategies.

Graduate Teaching Assistant

• Evaluated students' performance and offered feedback on their assignments, projects, and inquiries.
• Reviewed class materials, assignments, and labs for a group of students.
• Participated in weekly meetings with the professor to analyze students' progress.
• Course:INFO-H 516

Data Analyst

• Collaborated on designing and maintaining KPI-focused logistics dashboards.
• Conducted comprehensive data monitoring and quality audits, employing advanced visualization tools to ensure data integrity and operational accuracy.
• Streamlined task management processes by implementing JIRA workflows.

Associate Data Analyst

• Gathered and merged data from diverse sources into a standardized format, optimizing it for subsequent analysis.
• Engineered SQL data models, optimizing logistics data structures for enhanced analytical capabilities and system efficiency
• Processed and transformed large logistics datasets with over 100 columns from multiple sources
• Automated logistics data pipelines.

Indiana University Indianapolis

GPA: 3.94

Coursework:
• Deep Learning | Cloud Computing for Data Science
• Database Management | Data Visualization
• Informatics | Natural Language Processing
• Data Analytics | Statistical Inference

English to Telugu Machine Translation

Adapted the BERT Fused Model for parallel English-Telugu sentence pairs, achieving a BLEU Score of 26.75.

Designed and implemented preprocessing steps, including tokenization and normalization, tailored for linguistic peculiarities in the dataset.

Modified preprocessing scripts and normalization techniques, such as byte pair encoding, to suit Telugu language-specific characteristics.

Neural Machine Translation | Byte Pair Encoding | Language-Specific Preprocessing | BLEU Score

AI Persona Detector

This open-source tool is designed to detect whether text content has been generated by AI algorithms or authored by humans.

Deep Learning | Machine Learning

Cloud-Based Generalized Data Analysis Platform

Designed and implemented a highly scalable cloud application with AWS services for unlimited user scalability.

PySpark | Python | AWS Glue | AWS Lambda |AWS S3

Life Years Lost in the USA

Created an interactive website visualizing U.S. suicide trends, enhancing data engagement.

Plotly | Seaborn | Matplotlib | D3.js | Html

Cancer Management Database

Designed and developed a database system utilizing SQL and data modeling techniques to store and analyze 20 years of cancer treatment and patient data from CDC WONDER.

Excel | MySQL | Tableau

Job Matching Using Neural Network Approach

Built a Resume - Job Matching description algorithm with Neural Networks

PyTorch | Keras | Tensorflow |sklearn | NumPy | Transformers

Prediction of Survival of Patients in the ICU

identify and predict the survival of patients using demographic data, medical history and EHR (Electronic Health Record) records.

Achieved 97% accuracy through diverse classification algorithms, optimized Random Forest for an 8% boost, and employed SMOTE to balance the dataset for improved model performance

Scikit-learn | XGBoost | Pandas | NumPy | Matplotlib

Track A Tool

Collaborated with a team of 5 in building human activity recognition model using Tensor flow

Designed and built a Beagle Bone Black based device to track objects

Surveyed and collected dementia patient data from hospitals to understand their forgetting patterns