RUIMIN (CHERYL) WEI

New York, NY +1-929-***-**** adkuho@r.postjobfree.com www.linkedin.com/in/ruiminwei EDUCATION

FORDHAM UNIVERSITY, GABELLI SCHOOL OF BUSINESS New York, NY MS, Information Technology, GPA: 3.97/4.00 Sep 2019 - Dec 2020

● President of Fordham Chinese Business Society, Graduate Teaching Assistant for Business Communication

● Courses: Database Management, Data Mining, System Analysis and Design, Deep Machine Learning, Blockchain CHENGDU UNIVERSITY OF TECHNOLOGY Chengdu, CHN

BE, Internet of Things Engineering, GPA: 3.10/4.00 Sep 2015- June 2019

● Courses: Database Principles and Design, Data Structure, IoT Engineering Professional Training SKILLS

● Data Analysis: Python, SQL, Java; Tensorflow, MATLAB; AWS

● Business Intelligence Tool: Tableau; Alteryx; Excel;SAS; SPSS

● Speaking Language: Chinese(native), English(advanced) EXPERIENCE

THE GLIMPSE GROUP New York, NY

Data Science Intern Jan 2021 - Present

● Data Exploration:generated 10+test datasets through developing APIs calls from FRED including data, frequency, transformation, etc. variables; scrapped macro data from the ECB, BOJ, NBS (China) to create D6 VR visualizations.

● Software Testing: implemented 6+ categories of data(financial, retail, healthcare, etc.) to generate testing feedback reports; improved history data storage function to solve application shutdown problem.

● Product Development: combined 2D data report and 3D data trends visualization to improve the VR immersive experience.

CAIRE BEAUTY COMPANY New York, NY

Data Analyst Intern July 2020 - Nov 2020

● Data Analysis:retrieved 3GB sales transaction data from past 120 days and 1 million customer data from same level brands; proposed strategies for strengthening company’s compatibility from top 50% to 25%through vertical analysis.

● Customer Analysis: visualized target customer trends and feature correlations on Tableau; set a CNN model to classify customer feedback; optimized register process which increased new accounts by 10%.

● Product Development: speeded up 2 new series of product development based on analyzing customer consumption habits using the Association Rule model; designed 4+ marketing plan which helped the company's sales increase by 5%.

● Web Design: built and implemented a new online customer community website; generated testing report and achieved 20%increase in newsletter subscribers by improving discussion board and coupon function. INSTITUTE OF INTELLIGENT MACHINES, CHINESE ACADEMY OF SCIENCES Hefei, CHN Data Analyst Intern Nov 2018 - Jan 2019

● Project Development:Present the approach to the Intelligent Hospital; summarized hospital operation model from the case, sort out useful attributes and group them into 10+tables based on hospital departments.

● Database Design: draw the logical model diagram (E-R diagram); defined the primary key and foreign key of each table including the relationship between tables; mapped the E-R model into the relational model.

● Data Process:extracted historical data to the new database for data warehouse development(ETL); established a GAN model to generate randomized brain MRI images to solve CT data laxity problem.

● A/B Test: designed an A/B test by adding Intelligent Hospital as a floating advertisement which increased home page visits by 20%.

PROJECTS

SECOND PRIZE IN 2019 NYC EMERGING CASE COMPETITION New York, NY

● Scraped 1GB user data from the past 90 days of event behavior data, including events location, user engagement, official community number, futurist number for a new hub selection analysis using python.

● Bucket users into 6 groups based on user demographic data and generated 5+features through creating flows on Tableau to track the increased trends of new users and users’ participation.

● Implemented recent 5 years city’s unemployment rates, cost of living, and office space demand data in a two-layer stacked LSTM model to generate an estimated cost report of each candidate city(ex: Seattle, 4 million/year). CONSTRUCTION OF CREDIT CARD FRAUD PREDICTION MODEL Chengdu, CHN

● Applied an Auto-encoder neural network with L1 regularization to enhance the feature learning ability and reduce noise in high-dimensional credit card transactions data (2 million); handled imbalanced sample issue via SMOTE.

● Constructed a Logistic Regression classifier to predict customer fraudulency; applied leave-one-out cross-validation to get reliable results and hyperparameter tuning on sigmoid threshold, achieving 99.06% accuracy and 90% recall.