SAPNA MISHRA

*** ***** ****** ***** *******, IL *0601 312-***-**** adj85j@r.postjobfree.com LinkedIn GitHub Repository SUMMARY

Passionate about data driven decision making with more than 6 years of experience across analytics and data science. EDUCATION

M.Sc. The University of Chicago Chicago, IL Major – Analytics & Statistics Expected March 2021

Courses: Data Engineering Platforms, Linear Non-Linear Models, Data Mining Principles, Machine Learning, Time Series Analysis and Forecasting, Deep Learning & Image Recognition, Big Data Platforms Bachelor of Engineering in Computer Science SNDT, India Major - Computer Science July 2012

Courses: Database & Web Systems, Advanced Programming Concepts PROFESSIONAL / ANALYTICS EXPERIENCE

3M Virtual Internship Chicago Data Science Jun’20 - Sep’20 Building a Predictive Modeling Tool that would predict the Expected life of the indoor air filters:

• Identifying the most significant predictors affecting the filter life and build a Generalized Linear Regression Model to predict the expected filter life. This model prototype will help to improve the performance of the existing algorithm. Capstone Project University of Chicago Medicine School Chicago Computer Vision GitHub Link Feb’21 Determining the aggressiveness of Prostate Cancer using multiparametric MRI scans of patients

• Research on how building a U-Net CNN would help identify the aggressiveness of the cancer based on the MRI scans of the patients as an input. The proposed model will lead to an early and accurate detection of prostate cancer. Deep Learning Project University of Chicago Chicago Computer Vision GitHub Link Feb’20- May’20 Detect the Age, Gender & Race of a person in an Image

• Built 3 different CNN based models to detect the Age, Gender and Race comparing against pre-trained networks like VGGFACE2 and DeepFace.

Time Series Forecasting University of Chicago Chicago Time Series Models GitHub Link Feb’20- May’20 Forecasting the weekly total sales for Rossmann Drug Company across 1115 stores

• Demonstrated how Singular Value Decomposition (SVD) can be leveraged to reduce the number of ARIMA models from 1115 to 3 without significant loss in accuracy. The proposed approach minimized the overall compute time by reducing the number of time series predictions. Exploring opportunities to publish. LafargeHolcim, Data Analytics Mumbai, IND Analytics Expert Aug’17- Jan’19 Led a team of 2 members to inculcate an analytical mindset by solving various supply chain use cases under the direct guidance of the CIO. Pressure Point Analysis- Safety & Optimal Routing

• Using K-Means algorithm identified the most hazardous routes from the set of routes used for product transportation. These pressure points were flagged in the navigation app that alerted the truck drivers. TCS, Analytics & Insights Mumbai, IND Senior Business Analyst Oct’14- Aug’17 Retail Credit Model Redevelopment for BASEL II implementation - Mortgage and Home Equity portfolios

• Worked as an individual contributor responsible to consolidate historical account-level data.

• Performed Univariate and Bivariate Analysis to identify the key predictor variables that helped in predicting the default probability (PD), exposure at default (EAD) & loss given at default (LGD).

• The analysis assisted in getting rid of the outliers in the data and shortlist the most significant predictor variables for the BASEL II modeling process.

GI Group Mumbai, IND SAS Programmer Analyst (Deployed at HDFC Ergo) Jul’12- Oct’14 Predicting Fraudulent Motor Claims

• Developed and implemented Analytical base table housing the predictor variables by integration of heterogeneous data sources.

• Performed Univariate and Bivariate Analysis to identify the key attributes for the model development.

• Explored various models including Logistic Regression, Decision Tress and Random Forest.

• Based on the various accuracy metrics and accounting for model interpretability ; recommended the Logistic Regression based approach.

ACHIEVEMENTS & CONTRIBUTIONS

• Completed an online Data camp course - Data Scientist with Python track – Certificate #89574

• Completed an online training on Statistical Modelling in R.

• Fun SPOC at TCS responsible for organizing team building activities and team outings.

• Treasurer cum Sponsorship Head of Computer Society of India at the Under-Grad School. SKILLS AND LANGUAGES

Technical: R, Python, SAS, QlikView, Tableau, Power BI, Azure Databricks, Spark, SQL Server, Oracle, Teradata, MS Office, UNIX Certifications: Certified Base SAS Programmer (License –BP033655v9)

Contact this candidate