Resume

• 2016

  Master of Science (MS)

  Focused on Machine Learning

  Data analytics

  Visualization

  AI

  and Robotics

  Computer Science

  Specialization in Machine Learning

  Graduate Student Association - Member Focus on Machine Learning and Artificial Intelligence

  Georgia Institute of Technology

• 2010

  Doctor of Philosophy (PhD)

  Focused on data mining

  machine learning

  and computational optimization techniques.

  Industrial and Systems Engineering - Decision & Information & Analytics

  Graduate Student Organization

  SSIE

  President

  2012-2014 European Student Association

  Senator

  2010-2013 Alpha Pi Mu

  Associate Member

  2011-present

  Binghamton University

  Data Analytics and Visulization

  Machine Learning

  Analysis of Manufacturing Systems

  Foundation Of Adaptive Optimization

  Applied Soft Computing

  Advanced Simulation

  Advanced Human Factors

  Applied Multivariate Data Analysis

  Designing With Experiments

  Enterprise Systems Engineering

  Computational Photography

  Machine Learning for Trading

  Collective Dynamics

  Database Systems and Design

  Artificial Intelligence for Robotics

  Lean Six Sigma Black Belt Certification

  Dartmouth College

  Lean Six Sigma Green Belt Certification

  Dartmouth College

• 2007

  Master of Science (MS)

  Management Sciences and Quantitative Methods - School of Business

  Focused on dynamic programming and forecasting

  Dokuz Eylul University

• 2003

  Bachelor of Science (BS)

  Focused on operations research techniques and double majored in statistics.

  Double Major in Industrial and Systems Engineering and Statistics

  Industrial Engineering Association

  2003-2007

  Gazi University

• 000.

  In order to enable students to solve complex analytical problems using programming languages

  I updated the current statistical analytics course by incorporating R into the curriculum. This project was awarded $13

  The Kern Family Foundation

• 000

  Hanley Sustainability Institute Project Grant

  My three colleagues and I were awarded to design a framework for sustainable agricultural practices in urban areas. This project included numerous lab experiments as well as statistical analytics. We modeled listeria growth in several food items using mathematical models. This study was awarded $25

  Hanley Sustainability Institute

• R

  C#

  Machine Learning

  C

  Data Analysis

  Matlab

  SQL

  C++

  Supply Chain Management

  Mathematical Modeling

  Optimization

  Git

  Java

  Operations Research

  Python

  Statistics

  Design of Experiments

  Healthcare Management

  Github

  Data Mining

  Multi-stage Methodology to Detect Health Insurance Fraud

  Fraudulent behaviors have become a serious burden to insurance systems by bringing unnecessary costs. Insurance companies thus develop methods to identify fraud. This study proposes a new multistage data mining methodology for insurance companies to detect fraud. Stages utilize distance and density based clustering

  risk quantification

  and decision trees. This research decreases the cost due to insurance claim denials by 7%. Estimated savings are above 500

  000 annually.

  Multi-stage Methodology to Detect Health Insurance Fraud

  The objective of this study is to select the most predictive set of attributes to precisely classify the policy-holders

  regarding their health risk

  to offer a health insurance with an analogous premium. In this study

  the performance of several neural networks and a decision tree classification algorithm are examined and assessed to classify policy-holders regarding their health risk.

  A Data Mining Approach to Adjust Health Insurance Premium

  A network topology of all the blood distribution centers in a region is simulated

  and interactions among the centers are observed. Shortage rates demand satisfaction rates

  and average inventory levels are considered as system performance parameters

  and the best inventory levels are determined for each blood type. The project increases order fulfillment rates by 13% and decreases waste ratios by 7%

  A Real-Time Simulation Based Model for Blood Supply Chain Systems

  This study proposes a machine learning-based risk assessment tool for estimating breast cancer risk

  which has the potential to help healthy women be aware of possible risks associated with their current lifestyle and physical condition. Backpropagation (BP)

  Learning Vector Quantization (LVQ)

  Probabilistic Neural Networks (PNN)

  and a decision tree are applied to classify a population based on their risk factors.

  An Artificial Intelligence Approach for Breast Cancer Early Risk Assessment

  Marina

  Johnson

  BSH Home Appliances Group

  University of Dayton

  Binghamton University - United Health Services Hospital

  Comcast

  HUGO BOSS

  Montclair State University

  Drexel University's LeBow College of Business

  Greater Philadelphia Area

  -- Drexel University

  LeBow College of Business\n-- Teaching Python programming

  data analytics

  and R programming courses

  Adjunct Professor of Analytics

  Drexel University's LeBow College of Business

  Greater Philadelphia Area

  -- Comcast

  Enterprise Business Intelligence\n-- Senior Manager\n-- Responsibilities: Building both predictive and prescriptive models to increase customer acquisition

  upgrade

  and retention

  as well as to reduce churn and downgrade.\n-- Key Contributions:\nBuilt upgrade models to predict which subscribers are likely to add Xfinity Mobile to their current product line. These models are being used for marketing purposes. Marketers decide what customers they will contact using these models \n-- Prepared analytics lecture notes and will teach internal advanced data science course to Comcast employees using R

  Python

  PySpark

  H2O

  Hive

  and other platforms. This program aims to train the internal workforce to be the future \"citizen data scientist.\"\n-- Working on a Machine Learning Automation Project. This project will automize most of our predictive modeling process including preprocessing

  model selection

  and reporting.

  Principal Data Scientist

  Comcast

  Turkey

  -- Bosch Siemens Home Appliances \n-- Intern Consultant - Advanced Analytics - Bosch

  Ankara - Izmir

  Turkey\n-- Responsibilities: During my two-year part-time co-op

  I employed simulation and optimization methods to improve operations \n-- Key Contributions:\nPredicted demand fluctuations to optimize production plans. \nWorked on order cycle times and batch sizes. Implemented a dynamic batch size based production plan. Reduced lead time by 7%. Increased on-time delivery by 4%

  Operations Analytics Practitioner

  BSH Home Appliances Group

  Binghamton

  NY

  Research Associate - Data Analytics

  Binghamton University - United Health Services Hospital

  Montclair

  New Jersey

  Assistant Professor of Analytics

  Montclair State University

  Dayton

  Ohio Area

  -- University of Dayton

  Engineering Management & Systems\n-- Responsibilities: Teaching data analytics

  operations research

  and statistics courses as well as participating in research activities with industry partners \n-- Key Contributions:\nDeveloped a framework using data science methodologies (machine learning and optimization) to predict listeria growth in food items under several environmental and genetic conditions. This research is funded by Hanley Sustainability Institute.\nTaught statistics

  operations research

  and data analysis courses using R. \nBuilt mathematical models for a local hospital to predict and reduce hospital 30-day readmission.

  Assistant Professor

  University of Dayton

  Turkey

  -- Hugo Boss Textile and Fashion Industry\n-- Consultant - Operations Research & Advanced Analytics - Hugo Boss

  Izmir

  TR & Metzingen

  Germany

  2007 – 2010\n-- Responsibilities: utilized data analysis

  simulation and optimization to improve supply chain operations and decision making for one of the biggest clothing companies in the world \n-- Key Contributions:\nDeveloped optimization based frameworks for transportation and product delivery problems. This framework increased the DIFOT (Delivery in full on time) by 3%. \nClassified products based on K-means. The clusters were then used to build production cells. This reduced the set-up times and increased the productivity by 8%. \nDesigned a predictive model for pricing. This work provided different discount strategies that maximized the revenue.

  Sr. Operations Analytics Practitioner

  HUGO BOSS

  Member

  Institute of Operations Research and Management Science

  President

  Graduate Student Organization at Industrial and Systems Engineering Department

  Member

  Alpha Pi Mu

  Senator

  European Student Association

  Member

  Institute of Industrial And Systems Engineers