Python, SAS, R, SQL, Spark, AWS (Amazon Web Services), Databricks

JIAYU ZHANG

Email: *****.********@*****.*** Phone: 647-***-**** LinkedIn: linkedin.com/in/jz020 GitHub: github.com/wcdjiayuzhang Portfolio: https://bit.ly/2UDAHVt City: Toronto

SUMMARY

Results-driven data science professional with 5+ years of experience in statistical analysis and data modeling across the business process supported by strong analytical and communication skills. Proven talent for using data based analytical results to help companies improve marketing strategies, decision-making and problem solving to achieve a sales-driven result.

SKILLS

PROGRAMMING LANGUAGES: Python, SAS, R, SQL

BIG DATA: Spark, EC2, EMR, AWS (Amazon Web Services), Databricks, Hive, Hadoop

MACHINE LEARNING TOOLS: Scikit-Learn, Keras, Spark ML

VISUALIZATION: Tableau, Kibana, Matplotlib, Seaborn

DATABASE: MySQL, SQL Server, DynamoDB, Salesforce

CLIENT PROJECTS

Client: Samsung Canada, Call Center Dashboard and Adobe Analytics Jan. 2019 Current

Tools: SQL Server, Python, Excel Pivot Table and Pivot Chart, Adobe Analytics API

•Achieve SKU-optimization by building a dashboard in Excel with call center data stored in SQL Server and connected by Python;

•Delivered an analytical result to help improving marketing strategy and decision making by presenting the dashboard to the client;

•Provided extra data availability from Adobe Analytics API by identifying useful data and providing examples of use cases to enhance project process;

•Sharing knowledge and data using ideas with other teams to improve general performance.

•Lead a team to build and deploy a store-clustering model.

Client: iFuture Data Technology, Churn Model using Spark Machine Learning Nov. 2018 Jan. 2019

Tools: DataBricks, Spark ML, AWS S3, Seaborn

•Attended the weekly online meeting for a detailed look of the problem and updated process;

•Successfully proved existing over fitting problem in the model that the client was using and prepare for new model by connecting raw data in AWS S3.

•Offered possible solutions and customer behaviour reasoning based on the important features and built a better model with Spark modeling and raw data from AWS S3.

•Deliver the model and prepared a summary report, got commended by the project lead due to contributions.

Client: Beam Data Ltd, Google Analytics Dashboard using Python Mar. 2018 May 2018

Tools: Python, Excel, Google Analytics API, Tableau

•Meeting with client to understand the problem they were facing and find out data availability;

•Collected website behavioural data from Google Analytics Reporting API and customer behaviour data from Eventbrite API using Python;

•Analyzed usage patterns along different dimensions such as geography, user demographics, and interest category;

•Developed customized reporting dashboard for the marketing director and presented final results;

•Proposed SEO strategies for lead quality improvement.

EMPLOYMENT

Strategy Institute, Marketing Database Analyst Oct. 2017 Oct. 2018

•Delivered excellence in the analysis of qualitative data used in management reports and generated weekly KPI reports for sponsorship sales management team.

•Conducted relevant and current market research by performing searches on the web in industries such as Finance, Investment, Education, Technology, and Healthcare.

•Managed new client information including creating new accounts, uploading and updating the CRM Database (Salesforce), creating marketing segments based on particular rules for different products.

•Updated and created companies and contact accounts for existing marketing lists by validating data, removing bad, out-dated and irrelevant data, and filling empty fields in Salesforce.

•Audited data for compliance with established quality standards by following data program techniques.

DATA SCIENCE PROJECTS

Bank Loan Prediction Nov. 2018 Dec. 2018

Tools: MySQL, Python, Scikit-Learn, Seaborn, Matplotlib

Data source: Academic Institute

•Prepared table for modeling by creating a database and merging data from 8 tables;

•Built a connect between Python and MySQL database;

•Explore the data and prepare feature engineering for model fitting;

•Built a model to predict which customer could default on a loan, achieved an accuracy of 97%.

Sentiment Analysis Oct. 2018 Nov. 2019

Tools: Python, Scikit-Learn, Keras, Matplotlib

Data source: Kaggle.com

•Explored the dataset for understanding data distribution, the correlation between feature and label with visualizations;

•Built a base model with Scikit-Learn to find possible problems in the training dataset and come out ideas for further improvement;

•Increased 10% accuracy by building a Keras model with an embedding layer that records the word positions;

•Presentation: make a general conclusion of the model performance and methodologies then present the result to the instructor, assistant instructor, and classmates.

EDUCATION

University of Toronto

Sept. 2011 Apr. 2016

Honor Bachelor of Science

Double major in mathematics and statistics, minor in history

WeCloudData

Feb. 2017 Jan. 2019

Par-time Data Science Program

BrainStation

Apr. 2018 June 2018

Certificate Data Science 2018

SAS Certified Base Programmer for SAS 9

Nov. 2016

SAS Certified Advance Programmer for SAS 9 (Grade 93%)

Dec. 2016

Contact this candidate