Varalakshmi Vakkalagadda

GRADUATE STUDENT · GEORGE MASON UNIVERSITY

+1-646-***-**** adjlqg@r.postjobfree.com varalakshmi-vakkalagadda-461a8a18b/

Education

George Mason University Fairfax, Virginia

MASTER’S IN DATA ANALYTICS ENGINEERING Aug. 2019 - PRESENT

•Relevant Course work: Predictive Data Analytics, Natural Language Programming, Geo-Social Data Analytics, Big Data Essentials: Hadoop and spark Framework, Advance Health Data Mining, Applied Statistics and Visualizations.

Chaitanya Bharathi Institute of Technology INDIA

BACHELOR’S IN INFORMATION TECHNOLOGY Aug. 2015 – May. 2019

•Relevant Course work: Principles of Data Mining and Management, Design Analysis of Algorithms, Computational Intelligence, Data Structures (Analyzing complexity of Algorithms), Object Oriented Programming Languages.

Skills

Languages Python, C/C++, R, JAVA

Databases SQL (Relational Databases), NoSQL

Machine Learning XG Boost Decision Trees, Clustering, Bagging and Boosting, SVM, Random Forests, CNN, Keras, Tensorflow

Tools Microsoft Office (Excel, Power-point, Word), Tableau, AWS Services, Git, Power BI, Hive, Weka, Jupyter

Experience

George Mason University Fairfax, Virginia

GRADUATE RESEARCH ASSITANT Aug. 2020 – PRESENT

WIFI Contact Tracing

• The study uses WIFI data to create a decision support tool that provides ranked list of people that are potentially in contact with infected person.

• Increasingly complex approaches are used to predict location from enterprise-level WIFI data logs.

• The method predicting the movement of individuals based on WIFI access point locations and building floorplan (converted to graph) achieves best results. We also generated the simulated data to observe the potential contacts and generated statistical reports.

• We are also examining the movement patterns of individual movement from historical data to get more accurate results.

George Mason University Fairfax, Virginia

TEACHING ASSISTANT Sept. 2019 – Sept. 2020

•Responsible for teaching and guiding students in under-graduate courses Computational and Data Science, Introduction to Python and JAVA.

•Coordinated lab sessions in JAVA - Object Oriented Programming.

Samsung Research & Development Center Bengaluru, INDIA

SOFTWARE ANALYST Jun. 2018 - Jun. 2019

•Web scrapped the reviews of Employee's on development, culture and innovation on projects from company's internal portal. Using this large data we developed a tool that performs Textual Analytics using statistical novel based techniques to classify these reviews into sentiments.

•Applied Bayesian Networks, SVM and various unsupervised algorithms in training a model on structured data and used different evaluation metrics to identify the most suitable model.

•Applied TF-IDF, word2vec model, POS Tagging, Word Sense Disambiguation Pre-processing methods to accurately classify the results.

Projects

Predicting Severity of Accidents - US-Accidents dataset Classification and Visualization Analytics [ Github ]

•Trained and tuned different Machine Learning Models to predict the severity of the accidents and evaluated their performance using various metrics.

•Performed Co-relational analysis and generated various visualizations and interactive dashboards in Tableau and performed Hypothesis Testing.

•Utilized Natural Language Processing to understand interesting patterns in the description of the accident. [Pandas Numpy Nltk]

Predicting and Analyzing the Hospital Readmission Using MIMIC3 Dataset Machine Learning [ Github ]

•The main objective of project is to extract relevant features (select chorots) that are required to predict Readmission status of the patient.

•Analyzed the dataset to identify the top features by performing sensitivity analysis, Co-relational analysis and applying hyper parameterization.

•Trained various Machine Learning supervised Models like Random Forest, KNN, Logistic Regression, Decision Tree.

•Analyzed the ROC curves and AUC values to check for over-fitting of data. [Pandas scikit-learn Matplotlib Scipy Seaborn]

Geography of Taste using Yelp NLP [ Github ]

•Performed analysis on restaurant reviews to understand what kind of food and drinks are preferred by people in each city and there by predict the diversity of taste of a region.

•Related the results to identify if taste of food is a good indicator of socio-economic status and does people food preferences vary with their income level, race and age group. Worked to identify the trends and behavior of users.

•Processed the huge dataset by applying few NLP data manipulation techniques and there by extract features. Clustered the results to identify the important features of each cluster to understand the food preferences. [Word2Vec Spectral Clustering Cosine Similarity LDA

Certifications

AWS cloud Practitioner Certified Issued: Jul, 2020

Contact this candidate