Business Analyst

SAHIL RAINA *****.*****@****.***.*** LinkedIn +1-682-***-****

EXPERIENCE

The University of Texas at Arlington June 2016 – May 2017

Research Assistant II, Graduate Business Services Arlington, TEXAS

Uploaded and Maintained Database of Prospective and Current students of College of Business with over 50 daily updates.

Extracted data from Student Database of College of Business triggering SQL Queries, averaging over 25 queries a day with multi-table selections and joins, to generate ad-hoc reports.

Performed descriptive and predictive analysis, using various Statistical tool (SAS, Excel) on the extracted data for predicting future trend of applicant’s profile, and perform cost analysis to evaluate the finances allocations of various departments.

Forecast and visualize the data using Tableau & Excel. Build Dashboard and shared the finding to Dean and Directors.

Carried out internal and external marketing & recruitment drives for promotion of Graduate Programs offered by College of Business. Made Presentations using Prezi and MS PowerPoint for Information Sessions as a part of Marketing and Recruitments.

Mahindra Susten - E14 Technologies February 2014 – August 2015

Product Engineer, R&D - Mechanical and Automation Mumbai, INDIA

Lead the development of product, “Autonomous Electric Vehicle”, capable of cleaning Solar Power Facilities of 29 MW solar photo – voltaic plant.

Managed a branch office with responsibilities of managing workforce, expenses, handling procurement and logistics, organizing training sessions and assisting in recruiting & interviews

Conducted Market Research Analysis based on quantitative and qualitative parameters (Market Size, Market Trends, Market rate growth and profitability, Industry Cost Structure, Distribution channels, Key Success Factors) and Site Surveys to determine design and mechanics for the product.

Translated data from multiple sources into useful information using descriptive statistics and business drivers to be utilized by senior management for strategic decision making.

Patented the designs for lift and drag systems, used on the vehicle. Prioritized the prototype and developed it within 1/3rd of speculated time with only 20% of the allocated resources.

Future Generali India Insurance Co. June 2013 – February 2014

Service Engineer, Motor Claims & Underwriting Mumbai, INDIA

Accessed company's ERP system to download data of active customers, insured vehicles with sub categories, existing policies, agents, service centers and claim registering call centers.

Generated ad-hoc reports and monthly KPIs. Automated the generation of daily performance report, resulting in minor human errors and great accessibility to the team.

Using Excel, analyzed the extracted data on various attribute like Insured declared value, depreciation of insured parts, salvage value, claim size, claim settlement time, liability difference, taxes, servicing cost, policy plans, vehicle details, driver details, agent’s records, branch profits, etc.

Analyzed the existing policy plans and underwrote the changes by considering claim size, claim numbers, market value, depreciation and various other factors. Also by consulting various departments involved in operations for subject expertise. Helped raise region’s quarter profit by 6%

Adjusted more than 1000 claims and reduced the average claim amount by 15% by negotiating with the customer and in some cases by recognizing the fraudulent claims.

Achieved 100% on-time reporting and quality compliance rating that led to promotion in short time frame.

EDUCATION

Masters in Sciences in Information Systems - Business Analytics May 2017

University of Texas at Arlington, (GPA - 3.6) Arlington, TEXAS

Research: Development of algorithm for improving imbalance in minority class for Machine Learning.

Courses: Data Science, Advanced Business Statistics, Data Mining, Database Management, Engineering Economic Forecasting, Accounting Analysis, Data Warehousing & Business Intelligence

Bachelors in Engineering in Automobiles June 2013

University of Mumbai, (GPA - 3.7) Mumbai, INDIA

Courses: Operations Research, Vehicle Safety, Vehicle Dynamics, Automotive Technology

Non-Degree Seeking Certifications:

• R Programing & Data Science Toolbox from John Hopkins University on Coursera

•Python for Everybody from University of Michigan on Coursera

•Salesforce Administrator from Salesforce (In Progress)

•AWS Solution Architect Associate level from AWS (In Progress)

Skills & Technologies

•Data Analytics: Python (Pandas, NumPy, SciPy), R (mice, rpart, CARET), Google Analytics

•Statistical Analysis: MS Excel 2016 (Advanced), SPSS, SAS Enterprise, SAS Base

•DBMS & ERP: MS Access; MySQL, Oracle 10, SAP ERP (MM, PP, FICO)

•DW & Business Intelligence: SSIS, SAP BO, SAP BI, IBM Cognos BI v10, Tableau v10.1

•Project Management: M.S Project 2010, MS Visio

•Soft Skills: Communications, Negotiations, Presentation, Public Speaking

Master’s Research: Algorithm Development for Improving Minority Class Imbalance

Conducted research to develop an algorithm to improve minority class identification during training of algorithm.

Application: Credit Card fraud detection, Disease identification, and other fields where occurrence of an instance has low probability.

Reviewed over 30 literature surveys to study other scholar’s contribution in the field and to formulate the method and algorithm.

For validation, finalized 30 datasets from over 250 selected ones chosen from various sources, which matched the criteria of domains, quality factors, and having minority instances less than 20% of total instances.

Performed the experiment using existing techniques (SMOTE, oversampling, under-sampling) of improving imbalance and compared with results from our algorithm. Observed the improvement by over 30% on average and over 70 % in some cases.

Measured p value by performing T-test to justify the significance of our algorithm over other existing methods. The noted values of p are in range of 0.00001 to 0.007.

Academic Projects:

Statistical analysis of randomly sampled data using SAS Enterprise

The Data was regarding “Health Expenditures by State of Residence”, with attributes involving Region name, State name, Population in 1991, Population in 2009, Population growth percentage, Enrollment in Medicaid in 1991 & 2009.

Created Model based on Medicaid Enrollment in 1991 to test the Medicaid Enrollment for 2009.

Techniques involved include Hypothesis Testing, One Way/Two Way Analysis of Variance, T- Test, Least Square Equation, Tukey Test, Chi Square Test, and Multiple Linear Regressions including identification of Outliers, Influential Points and Multicollinearity.

Data Mining to analyze the dataset for Thyroid disease

Performed the design of experiment using Machine Learning classification algorithms, One R, Naïve Bayes, J48 and Random Forest to train the model to predict whether a person is suffering from Thyroid disease or not.

Applied various data cleaning and wrangling process and experiment design to improve the accuracy of data-set.

Developed a model using Knowledge flow feature in WEKA to draw multiple ROC curves.

Relational Database in MySQL

Designed production database maintenance schema. Created Tables, Views, Constrains, Synonyms & Sequences.

Worked with Relational Database Models, Schema, Entity-Relation diagrams to create technical/system design document. Created logical models, including E-R Diagrams and converted logical model into physical model.

Developed over 35 SQL Stored procedures, functions, triggers and views.

Financial Analysis of a company from investor’s perspective

Performed complete financial analysis using Financial Statements from 2008 to 2014, to analyze performance and growth of

“Johnson & Johnson” over last five years comparing with competitors, with an investment approach.

Used ratios include ROA, ROE, Gross Profit Margin, CR, QR, Asset Turnover, Inventory Turnover.

Social Media Sentiment Analysis using Twitter API

Performed sentiments analysis by collecting tweets tweets about Hilary, Trump and Obama and compared the sentiments of tweets of 2016 elections with 2008 elections.

Created Word-Cloud and calculated the percentage subjectivity and polarity of the tweets. Used twython, nlth, wordcloud.

Text Mining and Topic Modeling using Python and MongoDB

Created a tool, capable of Searching, Identifying and Classifying the duplicate or existing files on a Web Server.

Implemented nltk, Sklearn, Beautifulsoup, scipy, pymongo, matplotlib, numpy and pandas. Classified the data using K-Mean, Graph cluster, Hierarchical Clustering.

The tool was successfully tested on the website of “Boys Scout of America”, which had over 57000 files comprising of over 100GB of data located at over 15 servers.

Contact this candidate