Data Analyst, Data Scientist, Business Analyst

Mayur Bansal

Santa Clara, CA 669-***-**** ******@********.*** github.com/MB4511 LinkedIn Medium Education

Columbia University, New York Sept 2018 - Dec 2019 Master’s of Science in Applied Analytics GPA: 4.08 Relevant Coursework: Machine Learning (ML), Natural Language Processing (NLP), Storytelling with Data, Data Management, Deep Learning, Statistical Testing The Northcap University, Gurgaon Jul 2013 – May 2017 Bachelor’s in Civil and Environmental Engineering GPA: 3.55 Work Experience

Edifecs Bellevue, WA Data Science Intern May 2019 – Aug 2019

• Developed the ML app interface pipeline using pySpark by transforming highly unstructured datasets (10M+ transactions) into an integrated electronic medical record system (FHIR)

• Built a hybrid recommendation system based on patients’ claims data using the auto-encoder architecture and validated the model’s performance by applying NLP techniques in python.

• Led the data science team in generating sales leads by translating raw datasets into visually appealing dashboards to present an insightful story to the business clients. Byteflow Dynamics New York, NY Data Science Intern Nov 2018 – Apr 2019

• Developed machine learning models to predict the stock price performance of Fortune 500 companies based on the news articles published online.

• Processed textual data retrieved from financial articles on a weekly basis using R to extract the real time sentiment associated with the stock price of the company. Yellow Dove Overseas New Delhi, India Business Analyst Jul 2017 – Aug 2018

• Designed experiments for multiple A/B tests and built predictive models to increase the weighted distribution of superfood products by 25% in modern retail stores across India. Projects

Developed a recommender system using graph database for Lego Sets using Neo4j Feb 2020

• Leveraged the knowledge graph technology to build effective recommendations for Lego Validating user survey responses by analyzing mouse movement data Nov 2019

• Developed a ML framework to classify user survey responses as valid or non-valid by ensembling the predictions made by anomaly detection techniques, LSTM classifier and a Markov Model. De Black boxing the output of a KNN based hybrid recommendation system Aug 2019

• Decoded the method behind how the KNN algorithm assigns similarity scores to group patients which in turn augments how insurers target customers

• Implemented topic modeling using Latent Dirichlet Allocation (LDA) to group patients based on the distribution of diagnosis codes in their respective medical claim histories. How much for your Airbnb Ranked top 5% university wide Kaggle Competition Nov 2018

• Applied supervised learning algorithms such as Random Forest, boosting techniques (XGBoost) to predict the Airbnb rental prices in NYC.

Skills and Interests

Tools/Languages: Python (pandas, NumPy, scikit-learn, gensim), R (Shiny, dplyr, data.table), Spark

(pySpark, SparkSQL), Tableau, Neo4j, Advanced Excel, SQL, Deep Learning, Knowledge Graphs, Keras

Contact this candidate