Data science, Machine learning, programing, predictive modeling

MINA MORADI-KORDMAHALLEH

Charlotte, NC (***) *** - 1771 acz3ia@r.postjobfree.com https://www.linkedin.com/in/mmoradik

https://github.com/mmoradik

Professional Overview

Talented and self-motivated data scientist with significant experience in data mining, machine learning, programming, applied statistics and engineering.

Seeking a challenging opportunity to use excellent problem solving skills to automate processes and drive operational efficiency.

Skills and Tools

Machine Learning, Clustering, Classification, Optimization ● Big Data, Hadoop, MapReduce, Spark, Microsoft Azure

Statistical Modeling, Time Series Analysis, Regression ● SQL, MongoDB, Neo4j, Hive, Impala, Power BI, Linux Bash

Predictive Modeling, Feature Selection, Visualization ● Control Theory, Kalman Filtering, Signal Processing

Python, R, Matlab, Weka, Octave, MLlib, Text Mining ● Neural Networks, Deep Learning, Keras, Credit Risk Modeling

Education

5/2017 Ph.D. in Electrical and Computer Engineering, North Carolina A&T State University, USA

GPA: 4.0/4.0

Research Areas: Time Series Modeling, Machine Learning, Optimization

Thesis Title: Time Series Prediction with an Evolvable Partially Connected Recurrent Neural Network

2011 M.S. in Electrical Engineering, Control System, Isfahan University of Technology, Iran

GPA: 3.9/4.0

Research Areas: Sensor Fusion, Statistical Modeling, Hypothesis testing

Thesis Title: A Fault-Tolerant Distributed Detection of Simultaneous Events in Wireless Sensor Network

2008 B.S. in Electrical Engineering, Control System, Imam Khomeini International University, Iran

GPA: 3.4/4.0

Thesis Title: Simulation and Design of Half Bridge and Full Bridge DC-DC Converter

Professional Experience

05/2016-08/2016 Intern Data Scientist, Analytic Team, MaxPoint, Morrisville, NC

ETL and data munging for real-time bidding in online advertising (Impala, SQL, Pandas)

Developed a mixture of expert LSTMs (Long Short Term Memory Recurrent Neural Networks) model for indoor localization and tracking of mobile wireless devices using their received signal strengths (Numpy, Keras, scikit-learn)

01/2014-05/2017 Graduate Research Assistant, North Carolina A&T State University, Greensboro, NC

ETL and predictive modeling of big data (Spark, Mlib, Microsoft Azure)

Developed a sparse recurrent neural network trained with heuristic optimization for predictive modeling of time series. Achieved 30% improvement in prediction of US dollar/British Pound exchange rate and 5% improvement in stock market prediction of Microsoft corporation. Achieved a high accuracy in prediction of hurricane trajectory using dynamic time warping and sparse neural networks. (Python, Pandas)

Parameters tuning of the sparse neural network with a bi-level metaheuristic optimization and random forest regression (Matlab)

Developed a tree-based hierarchical recurrent neural network for finding the causal interactions in biological networks (R, Python, NetworkX)

Developed a hierarchical multi-label classification technique for protein function classification (Weka)

Modeling of probability of default of bank customers using their credit card billing and payment history, behavioral information and loan status (Python)

Cleaning, visualization and comparisons of the Google and Amazon reviews using the bag of words text mining technique (R)

Cleaning of the Titanic passenger data for purpose of classification and prediction of the survival using supervised machine learning techniques (Python)

Modeling of non-stationary time series with change points using regression and convex optimization

09/2009-09/2011 Research Assistant, Isfahan University of Technology, Isfahan, Iran

Developed a distributed fault tolerant event detection algorithm in wireless sensor networks based on Bayesian statistics, likelihood ratio tests and Kalman Filtering

Teaching courses for undergraduate level on linear control systems

05/2008-08/2008 Engineering Intern, Gilan Power Distribution, Rasht, Iran.

Collaborated with engineers and project managers regarding design parameters for electrical power distribution and lighting

Certificates

Statistical Modelling in R (Part 1 &2), Statistical Thinking in Python (Part 1 &2), Introduction to Statistics with R: Student’s T-test, Text Mining: Bag of Words, Credit Risk Modeling in R, ARIMA Modeling with R, Correlation and Regression from DataCamp.

Honors and Awards

2016 First winner award, 5th annual graduate student research competition, NCAT

2015 Travel award, 5th international workshop on climate informatics, NCAR

2016-2017 Graduate scholarship, expeditions in computing, research on understanding climate change, NSF

2014-2015 Interdisciplinary scholarship award, the study of evolution of action (BEACON), NSF

2013 Teaching assistantship, department of electrical and computer engineering, NCAT

Professional Activities

Membership: IEEE Student, IEEE Women in Engineering Reviewer: Journal of Neural Computing and Applications

Selected Publications

Journal Papers and Book Chapter:

•M. Moradi, M. Gorji, A. Homaifar, “Application of a novel partially connected artificial neural network with evolvable topology in time series prediction”, Applied Intelligence, submitted.

•M. Moradi, M. Gorji, H, Scott, A. Homaifar, “Identifying time-delayed gene regulatory networks via an evolvable hierarchical recurrent neural network”, BioData Mining, submitted.

•M. Gorji, M. Moradi, A. Homaifar, S. Liess. “Change detection in climate time series based on bounded-variation clustering”, Machine Learning and Data Mining Approaches to Climate Science, pp. 185-194, Springer, 2015.

Conference Papers:

•M. Moradi, M. Gorji, A. Homaifar, “A sparse recurrent neural network for trajectory prediction of Atlantic hurricanes”, Genetic and Evolutionary Computation Conference (GECCO), pp. 957-964, Denver, CO, 2016.

•M. Moradi, M. Gorji, A. Homaifar, “A Bilevel Parameter Tuning Strategy of Partially Connected ANNs”, IEEE 14th International Conference on Machine Learning and Applications (ICMLA), pp. 793-798, Miami, FL, 2015.

•M. Moradi, M. Gorji, A. Homaifar, A. Karimoddini, A. Guiseppi-Elie, J.L Graves “Delayed and hidden variables interactions in gene regulatory networks”, IEEE 14th International Conference on Bioinformatics and Bioengineering (BIBE), pp. 23-29, Boca Raton, FL, 2014.

•M. Moradi, M. Gorji, A. Homaifar, D. KC “Time-series forecasting with evolvable partially connected artificial neural network”, Genetic and Evolutionary Computation Conference (GECCO), pp. 79-80, Vancouver, BC, Canada, 2014.

•M. Moradi, A. Homaifar, D. KC, “Hierarchical multi-label gene function prediction using adaptive mutation in crowding niching”, IEEE 13th International Conference on Bioinformatics and Bioengineering (BIBE), pp. 1-6, Greece, 2013.

•M. Gorji, M. Moradi, A. Homaifar. ‘’Identification of switched models in non-stationary time series based on coordinate-descent and genetic algorithm’’, Genetic and Evolutionary Computation Conference (GECCO), pp. 1399-1400, Madrid, Spain, 2015.

•M. Gorji, M. Moradi, A. Homaifar, A., Karimoddini, “Switched linear system identification based on bounded-switching clustering”, American Control Conference (ACC), pp. 1806-1811, Chicago, IL, 2015.

•M. Moradi, M. Gorji, J. Ghaisari, J, Askari, “A new method for detection of a distributed event in wireless sensor network”, 19th Iranian Conference of Electrical Engineering (ICEE), pp. 1-5, Tehran, Iran, 2011.

Relevant Courses

•Computational Statistics, Probability Theory, Numerical Methods, Linear Algebra, Engineering Mathematics, Applied Machine Learning, Econometrics

•Artificial Neural Network, Genetic Algorithms, Pattern Recognition, Image Processing, Introduction to Bioinformatics

•Optimal Control, Nonlinear Control, Robust Control, Digital Control, Multi-Variable Control Systems

•Industrial control, Instrumentation, Power Electronic, Electrical Machines, Siemens PLCs Programming

Contact this candidate