Data Analyst

Michael (Hau-an) Shieh

206-***-**** ********@**.*** Seattle, WA Linkedin: mikesh13 Portfolio Advanced data modeling and analysis skills for solving problems with solid understanding of machine learning SKILLS

- - - Proficient Effective Data Science in with Hypothesis Libraries: R, Python, lme4, Testing, and glmnet, Relational Regression FactorMineR, Database models, mclust, Clustering, (SQL) and randomforest, Familiar Dimensionality with tidyverse, PHP, Reduction, HTML, ggplot2; C+Decision +Pandas,, Jupyter, Tree, NumPy, Docker, and SciPy, Design and Matplotlib, PowerBI of Experiments Seaborn, scikit-learn, TensorFlow, Keras

EDUCATION

University of Washington, Seattle, WA, United States Sep. 2018 — Aug. 2020 Master - - Research Relevant of Science, Coursework: Topic: Security Industrial Data Update Analytics, Engineering Decision Data Making Programming, with Positive Inferential Reinforcement Data Analysis, Statistical Learning, Design of Experiments,

- DubsTech Stochastic Data Processes, Science Optimization Certificate National Cheng Kung University, Tainan, Taiwan Sep. 2012 — Jun. 2016 Bachelor of Science, Management Science

- - Research Relevant Coursework: Topic: Comparisons Applied Between Statistics, Regular Operations and Economical Research, Decision-Screening making Processes Methods, Using Database Simulation Management Methods EXPERIENCE

Data Analyst Intern, Golden Financial Exploration Association, Bellevue, WA, United States Jun. 2020 — Aug. 2020

- Built and maintained MySQL database for data collections and a Python program to importing data from Google Drive to solve the

- - - problem Provided Processed Modified that statistical survey and data visualized questions were consultation scattered data that weekly resulted for around analyzing, to in present multiple responses modeling, findings spreadsheets with with and better predicting internal quality stakeholders to trends simplify using the for the subsequent improving feedback future data collected processing events from event participants Teaching Assistant, University of Washington, Seattle, WA, United States Jan. 2019 — Mar. 2020

- - - Probability Tutored Improved undergraduate students’ and Statistics performances students for Engineers during by initiating (1 weekly quarter)conversations office ; Manufacturing hours and with assisted Scheduling students professors to and collected Inventory in constructing feedbacks (1 quarter)regularly homework ; Simulation to help and (instructors exam 2 quarters) questions adjust materials, pace in classes, etc.

Quantitative Researcher, Institute of Information Science, Academia Sinica, Taipei, Taiwan Jan. 2018 — July 2018

- - - - Developed Processed Yielded Collaborated 50% trust investment more with fund fellow return and strategies researchers compared S&P 500 by data predicting to by historical preparing for data stock average preparation data market for return research performances and annually eliminated projects, using missing reviewing regression value progresses, with models KNN and and algorithm deep exchanging learning ideas models DATA SCIENCE PROJECTS

Analyzing the Relationships between Airbnb Listing Prices and New York City Reported Crime

- - - Aimed Processed Found to average understand and visualized listing if prices local current and crime occupancy listings rate affects and rates reported customers’ are higher crimes preferences in over boroughs the on past with Airbnb 20 lower years listings, crime by categories and rate consequently, except to analyzed for Manhattan affects crime listing rate trends prices. Security Update Decision Making with Positive Reinforcement

- Improved existing online experiment software to identify key features that affect humans’ learning behavior on making decisions using

- - design Collected Found positive of experiments and stored reinforcement data techniques in SQL have database little effect on on SSH improving server for human the subsequent behaviors analysis using regression models using RStudio Predicting Growth of Gross Domestic Product per Capita

- - - - Intended Cross-Implemented Raised validated correlation to predict dimensionality different by GDP 33% growth machine and reduction reduced using learning human techniques learning models development error to to select select by 10% 44 the indexes after independent best clustering from approach the variables World the by countries comparing Bank from HDI over by learning dataset economic 1600 variables error using indicators RStudio to avoid overfitting

Contact this candidate