Post Job Free

Resume

Sign in

Data Driver

Location:
San Jose, CA
Posted:
March 20, 2019

Contact this candidate

Resume:

XINGWEI WU

Email: ac8ua9@r.postjobfree.com LinkedIn: https://www.linkedin.com/in/xingwei-wu/

Phone: 206-***-**** Address: 1461 Lee Pl, San Jose, CA, 95131 SUMMARY

● Expertise in statistical modeling, inference and predictive analytics with real-world applications

● Extensive hands on experience in behavioral studies, experimental design, data analysis, and visualization

● Over 6 years of research experience as demonstrated in international journals and conferences

● A detail-oriented person, quick learner, problem solver, and a great communicator in teamwork settings EDUCATION

● Ph.D. in Transportation Engineering, University of Washington 2013 - 2019 Expected Advisor: Dr. Linda Ng Boyle

Dissertation: Understanding the Effects of Auditory Cues for V2V- based Crash Warning System

● M.S in Statistics (Data Science Track), University of Washington 2015 - 2018

● M.S. in Transportation Engineering, Tongji University, Shanghai, China 2010 - 2013

● B.S. in Transportation Engineering, Tongji University, Shanghai, China 2006 - 2010 CORE COURSES:

● Machine Learning for Big Data ● Data Structures and Algorithms

● Foundations of Machine Learning ● Data Management

● Statistical Inference ● Design and Analysis of Experiments

● Applied Regression ● Statistical Computing

● Bayesian Statistical Methods ● Stochastic Modeling of Scientific Data

● Analysis of Categorical and Count Data ● Sample Survey Techniques SKILLS

● Programming languages: R, Java, Python, SQL

● Statistics: A/B testing, Analysis of Variance techniques, Prediction Modeling, Clustering & Classification

● Analytical modeling: R, SPSS, SAS, WinBUGS

● Big data ecosystem tools: Hadoop, MapReduce, Spark SELECTED ACADEMIC PROJECTS

Diagnose Breast Cancer from Fine-needle Aspirates Using Machine Learning

● Dataset consists of the diagnosis and features describing the cell nuclei present in the digitized image of a fine-needle aspirate (FNA) of a breast mass

● Feature selection using PCA analysis

● Constructed Logistic Regression, Random Forest, Adaboost and Gradient Boosting Machine (GBM) models to predict whether a breast cancer is benign or malignant.

Predicting Movie Review Sentiment using Naive Bayes

● Dataset consists of labeled 50,000 IMDB movie reviews, specially selected for sentiment analysis

● Encoded reviews with the bag-of-words model representation

● Built a Naive Bayes classifier to predict encoded movie reviews as positive or negative Housing Price Prediction Using Machine Learning Algorithms: The Case Of King County

● Collected and synthesized 10-year housing data in King County, WA

● Built Support Vector Machine and Neural Network to predict exact unit price and price range using Python Netflix Movie Recommender System using MapReduce in Hadoop

● Designed a recommendation system based on users’ past preference and movie rating histories based on users’ historical rating data from Netflix

● Implemented Item Collaborative Filtering algorithm, built co-occurrence matrix and rating matrix, and generated recommender feeds.

PROFESSIONAL EXPERIENCE

Researcher, Human Factors and Statistical Modeling Lab, University of Washington 2013 – present Industrial & Systems Engineering, UW

Vehicle-to Vehicle Safety Application (Sponsored by US DOT) Analyzed drivers’ crash avoidance behavior and evaluated in-vehicle auditory crash warning system (CWS)

● Developed pre-crash event scenarios and in-vehicle auditory CWS for a simulator driving study

● Designed and implemented driving simulator experiments with more than 400 subjects for three different CWSs

● Implemented algorithms for data reduction, selection and quality assurance using MATLAB

● Used repeated measures ANOVA and regression models to quantify the relationship between crashes and driver behavior

● Applied Partial Least Square Path Modeling to predict drivers’ response to alert and crash event outcome

● Analyzed drivers’ perceptions and attitudes towards CWS using Mixed Effects Regressions Examining the Effectiveness of Entry Level Driver Training in Improving CMV Driver Safety Performance Examined Commercial Driver Licenses (CDL) training effectiveness based on crash and violation records over 10 years

● Data reduction and cleaning for over 90,000 CMV driver and 300,000 violation data using SQL and R

● Developed a Seasonal ARIMA time series model and Intervention analysis to assess the effectiveness of CDL training based on longitudinal violation data with over 10 years of Washington state data Evaluation of US Carrier Safety Measurement System (Sponsored by NAS) Evaluated the implementation of the Safety Measurement System (SMS) Methodology using carrier group and geographical factors

● Developed data visualization and cluster analysis on crash and violation records to identify safety performance for different carrier groups

● Developed hierarchical models to predict crash involvement and violation occurrence Statistical Consultant, University of Washington Jan. 2017 - Apr. 2017 Held consulting sessions to answer questions, provided statistical suggestions and solutions and followed up with clients

● Synthesized information, answer the clients’ design or analysis problems

● Prepared statistical considerations for experimental designs and data analyses

● Structured assumptions, references and potential limitations of the suggested approach

● Provided bespoke statistical analyses tailored to the requirements of the clients SELECTED PUBLICATIONS AND PRESENTATIONS

● Wu X., Boyle L.N., Marshall D., O’Brien W. (2018). Evaluating the Effectiveness of Auditory In-Vehicle Forward Collision Warning Alerts Based on PLS Path Models. Transportation Research Part F: Traffic Psychology and Behaviour, Vol (59), 164-178

● Wu X., Boyle L.N. (2019). Auditory Warning Messages Within an Intersection Movement Assist (IMA) System: Effects of Speech and Non-speech Based Cues. Human Factors: The Journal of the Human Factors and Ergonomics Society (under review)

● Wu X., Guo H., Boyle L.N. (2019). The impact of Washington's commercial driver training program on driver safety. Accident Analysis & Prevention (under review)

● Wu X., Boyle L.N., Marshall D. (2017). Drivers’ avoidance strategies when using a Forward Collision Warning (FCW) system. In Proceedings of the Human Factors Society 61th Annual Meeting, Austin, TX

● Wu X., Guo H., Boyle L.N. (2017). Evaluating Safety Performance Measures for Motor Carriers Using Hierarchical Models. Presentation at the 2017 Joint Statistical Meetings, Baltimore, MD

● Wu X., Boyle L.N. (2015). Sampling Biases Associated with Driver Distraction Tasks in a Simulated Environment. In Proceedings of the Human Factors Society 59th Annual Meeting, Los Angeles, CA.

● Wu X., Guo H., Boyle L.N. (2015). The Effectiveness of Commercial Driver Training: A Times Series Modeling Approach. Presentation at the 2015 Joint Statistical Meetings, Seattle, WA

● Wang X., Wu X., Abdel-Aty M., Tremont P. (2013). Investigation of Road Network Features and Safety Performance. Accident Analysis & Prevention, 56, 22-31



Contact this candidate