Yizhuo Li

Data Scientist, Lifelong learner

sit

About Me

Hello, this is Yizhuo Li. I am a life long learner Data Scientist.

This year, 2020, I will graduate from UT Austin, Information Studies program.

Hope you have fun with this website.

Data Science

About my Internship, project and Capstone

VIDEOS

Videos that my teammember or I introduce our projects

PHOTOS

Photos by me

Want to see more? Subscribe to my Youtube Channel!

DS Intern at Grid Dynamics

• Developed fraud detection models for online transactions and predict fraud score for transactions. • Deployed predicting models to AWS, and automatically retrain and update models with new coming data on a regular basis, building APIs for backend access. • Developed ARIMA and Prophet time series models to predict future loss, also automatically update with the latest data.

Research Assistant of Data Analyst

• Grabbed millions of Baidu academic literature information using Selenium and use multi-threading to improve efficiency. • Collated the last 5 years of 2000+ organizations’ report data (100+ fields) from 3 different data sources by Python. • ETL raw unstructured data into actionable large dataset for next steps. • Built the standard and compliant data pipeline and use Git to finish the version control with other team members.

I am going to add some new features to the old web system of the Texas State Board of Pharmacy (TSBP) which makes the whole system more automatic. Currently, the TSBP will receive the orders from the wholesalers by Email or submit a specific type of file from TSBP’s unique system. However, the problem is the system now can’t automatically load the data from the uploaded file and save the data into the TWDD database. Currently, this step is finished by a technical employee. Also, although the file’s type is required, the wholesalers sometimes will not follow the requirements. So, I try to make the whole process automatic and check whether the uploaded files are in the ARCOS format. If I finish these aims, that will help the TSBP to improve their efficiency.

Project Activities and Methods

  1. Validate files uploaded to a public server by wholesalers are in ARCOS format

  2. Inform wholesaler of file upload status (success/fail)

  3. Transfer files from a publicly accessible server to internal TWDD database server via FTP

  4. Import valid files into the TWDD database.

  5. Provide import data reporting to staff

 

Project Deliverables

  1. A system schema for the automated system.

  2. A web system builds by Larval that implements this schema.

  3. Several user test cases of the system.

  4. Documentation for the system.

  • (737) 205-0588
  • isyizhuo.li@gmail.com
  • 1600 Wickersham Ln, 3012; Austin, TX; 78741

I typically reply to all email inquiries within 24 hours.