ABBAS HOOSHMAND, PHD

actca2@r.postjobfree.com, 206-***-****, Seattle, WA

SUMMARY OF QULIFICATIONS

• Data Science: Machine learning, data mining, statistical & predictive modeling

• Programming: Python, Java, C# (in .NET), Fortran, Bash, Pascal

• Biometrics: Experienced in HRV and sleep analysis

• Softwares: SQL, Hadoop, Pig, Eclipse, R, Octave, Matlab, LATEX, Microsoft Office

• Data Visualization: R ggplot2 and Python Matplotlib libraries

• Operating Systems: Mac OS, Linux, Windows

JOB EXPERIENCE

• Wave modeler and data scientist intern at USGS, Apr 2015 - Dec 2015 Derived an analytical model to investigate time-series tide data for predicting sea level rise and vertical land movement

• Model Developer Intern at RMS, Jun 2014 - Sep 2014 Applied Machine Learning techniques using Python to update US river network information with comparing model output and observational data

– Derived data from satellite images and Google Earth API using Python

– Performed statistical analysis on flood inundations using ArcGIS, GRASS, OGR

– Updated levee and defense information using Sensitivity and Precision

– Applied the developed methodology to other data sets and verified its performance

• Research Assistant at University of Washington, Sep 2009 - Mar 2015 Developed an analytical and numerical model using R and MATLAB to predict surface wave energy and performed statistical analysis on experimental results EDUCATION

University of Washington, Seattle, WA

MS/PhD in Civil and Environmental Engineering, Dec 2014 Minor in Statistics and Machine Learning

Sharif University of Technology, Tehran, Iran

BS in Mechanical Engineering, Mar 2009

CERTIFICATES

University of Washington, Seattle, WA

Certification in Data Science, Expected Mar 2016

Content: Text Analysis, Parallel RDBMS, Storage and Concurrency Preliminaries, Bayesian Statistics, Testing and Experimental Design, Graph Algorithms Classification, Dimensionality Reduction SELECTED PROJECTS

• Sentiment analysis based on random walk in Twitter, Sep - Dec 2012 Compared social network sampling methods on Twitter and performed sentiment analysis on sampled tweets.

– Applied three sampling techniques to choose random Twitter users including unadjusted random walks over users and random walks with Metropolis-Hasting correction.

– Implemented convergence diagnostics on random walks of users.

– Ran sentiment analysis on sampled tweets of user profile data using two different methods: Naive Bayes classification technique and word polarity scoring

• Prediction of mortality due to Osteoporosis in males, Jan - Mar 2013 Applied multi-variable regression on longitudinal data set of 2000 biomarkers to forecast mortality in male osteoporosis cases.

– Analyzed statistical distributions of predictive variables

– Used Bayesian Model Averaging (BMA) to choose predictive features

– Used Alternating Conditional Expectations (ACE) to determine possible transformations on variables

– Used Multivariate Imputation by Chained Equations (MICE) to fill in missing data.