Experience in Python, SQL skills

Hsuan-Ju Lin

** **** ******, *** ** Hoboken, NJ *7030

******@*******.*** (201-***-*****

EDUCATION

Stevens Institute of Technology

Hoboken, NJ Cumulative GPA: 3.834

Master of Science in Business Intelligence & Analytics, Specialization: Data analytics

Developed SVM model using Natural Language Processing in Python to predict crowdfunding outcome

Predicted Lending Club loan default by utilizing different machine learning methods in a team of four

Participated in Portfolio Optimization project by applying MATLAB and Excel Solver to model and optimize Sharpe ratio

Led design of experiments project to analyze customer’s purchase decisions by leveraging JMP with survey TECHNICAL SKILLS

Experienced in Interactive Data Visualization: Python (Seaborn, Matplotlib), R (ggplot), Tableau

Experienced in Explosive, Descriptive, Predictive Data Analysis: Excel Advanced Functions, Python

(scikit-learn, nltk), R (neuralnet, kknn, class, C50)

Experienced in Mathematical and Statistical Modeling: SAS

Experienced in Programming Languages: SQL, Python, R

Experienced in Hadoop Big Data tools: HDFS, MapReduce, AWS

Experienced in Agile Scrum and Waterfall methodologies in Software Development Life Cycle EXPERIENCE

The Pennsylvania Market Restaurant

Pittsburg, PA

Marketing Analyst Sep 2019 – Dec 2019

Identified business problems with restaurant manager by performing descriptive analysis on transaction data, online Yelp and google reviews and demographics

Collaborated with restaurant manager to perform conjoint analysis utilizing SAS with questionnaire in addressing lunchtime slump issue and bad impression for food found in descriptive analysis

Applied Holt-Winters additive and multiplicative seasonality time series algorithm to forecast the sales of the following 30 days using SAS to help manager with decision making in a team of four Stevens Institute of Technology

Hoboken, NJ

Research Assistant Jun 2019 – Nov 2019

Gathered the requirements from professors to identify the modeling requirements in evaluating the role of project novelty in initial coin offerings mechanism for bitcoin technology

Crawled the data from ICObench and ingested the data in Python to support the modeling requirements

Evaluated usefulness k-medoid algorithm, LDA, and self-organizing map method for clustering, but concluded these methods were insignificant

Verified novelty is negatively associated with the amount raised and rating by carrying out a series of ordinary least squares regression models with robust standard errors in a team of six

Contributed to paper titled "EVALUATING NOVELTY AT INITIAL COIN OFFERINGS", accepted for presentation at The 2019 Pre-ICIS SIGBPS Workshop on blockchain and smart contract

Contact this candidate