Ahmet Cecen

Programmer / Data Scientist

PhD

F-1

OPT

500 10th Street NW
Atlanta, Georgia GA 30332 US
(267) 586-4505
ahmetcecen@gmail.com
http://ahmetcecen.github.io/

Versatile programmer looking for a full time position around Summer/Fall 2017. Uncanny ability and enthusiasm to assimilate, adapt and build on existing code and concepts in pretty much any language and project.

Toolsets

Machine Learning & Data Analysis
  • Image/Signal Processing
  • Convolutional Neural Networks
  • Natural Language Processing
MATLAB R Python
Web Development
  • Front End Design
  • Visualization
JQuery D3.js Three.js
Deployment & Automation
  • Software Prototyping
  • Hardware Prototyping
  • Lab/Workflow Automation
Batch Arduino Labview
HPC & Big Data
  • Map-Reduce
  • MPI
C++ Spark Hadoop

Education

  • Georgia Institute of Technology

    Doctorate08/2013 - Present

    Computational Science and Engineering

    Thesis: Calculation, Utilization, and Inference of n-point Statistics in Practical Spatio-Temporal Data

  • Drexel University

    Bachelors09/2009 - 08/2013

    Mechanical Engineering and Mechanics

    Graduation Project: Design and Manufacturing of an Automated Flow Capacitor Test Station

Work Experience

  • Georgia Tech - MINED Group

    Graduate Research Assistant

    03/2010 - Present

    • Design, implement and publish efficient machine learning and informatics tools for processing, statistical analysis and visualization of 2D/3D/4D image data.

  • Georgia Tech - Melkers Group

    Graduate Research Assistant

    01/2014 - 06/2014

    • Tailor big data analysis and machine learning tools to develop filters that resolve name disambiguation in the web of science publication index.

Selected Publications

Efficient and Accurate Materials Structure-Property Linkages using 3D Convolutional Neural Networks, KDD2017 - Submitted
Role of materials data science and informatics in accelerated materials innovation, MRS Bulletin - 2016
Versatile algorithms for the computation of 2-point spatial correlations in quantifying material structure, Integrating Materials and Manufacturing Innovation - 2016
Exploration of data science techniques to predict fatigue strength of steel from composition and processing parameters, Integrating Materials and Manufacturing Innovation - 2014

Community Impact

Google Scholar Citations
Citations 205
h-index 6
i10-index 6
MATLAB Central
ANSWERS105RANKof109,937
FILE EXCHANGE590RANKof12,160
GitHub Language Usage

Awards

  • Air Force Research Lab in partnership with NIST and NSF

    Materials Science and Engineering Data Challenge - Runner Up05/2016

Projects

18 88

Collaborative blogging for the domain expert and the data scientist.

Javascript HTML/CSS MATLAB

Python XSLT Jekyll

Downloads: 10,000 (est.)

Polynomial regression with cross-validation on multidimensional data.

Machine learning tool for multiscale materials science.

Python Batch

Online demo written in R using Shiny framework.

R HTML/CSS

Computes 3D spatial statistics of gigantic datasets via memory mapping.