Sven Dong 🧋

Sven Dong

Master of Data Science @UBC

University of British Columbia

Hello, I am Sven (Xinyu), currently studying for a Master of Data Science at UBC. With a focus on large data modelling and visualization, I possess valuable expertise in this field. My professional experience includes working at Deloitte and Daimler, where I honed my skills and gained practical knowledge in the industry. I am passionate about leveraging data to extract meaningful insights and drive informed decision-making.

Skills

R

90%

Statistics

100%

SQL

70%

Visulization

90%

Python

70%

Communication

70%

Experience

 
 
 
 
 
Siemens Healthineers
Data Scientist(Capstone)
May 2023 – June 2023 Oakville

Responsibilities include:

  • Supervised Modelling for Downtime: Utilized tree-based method(Random Forest, Gradient boosting) to model the production line logs, achieving statistical inference and factor analysis for downtime.
  • Text Mining: Applied non-parametric clustering to the textual logs in the production line, identifying latent patterns. Visualized the clustering results and work closely with the client to interpret them effectively.
 
 
 
 
 
Mercedes Benz Financial Service, Daimler
Data Analyst Intern
July 2021 – September 2021 Beijing

Responsibilities include:

  • Database Analysis: Tested database dependency and simulated interface files for ETL implementation in the data warehouse environment. Defined the data stream of individual information in ERPS to comply with GDPR.
  • SQL Development Review: Performed preliminary SQL code reviews for different data centres, and drafted the correspondent project documentation on Confluence and JIRA.
 
 
 
 
 
Deloitte
Business Analyst Intern
January 2021 – March 2021 Beijing
  • Business Consulting: Assisted in strategic consulting for five dealers of Volkswagen. Communicated with managers from different business aspects to investigate the operating status and supply chain management. Experienced with communicating and identifying stakeholder needs.
  • Integration and Data Preparation: Established a pipeline to collect, clean, integrate and validate data from OMS(Order Management System) and related financial footnotes to an organized clean data source. Familiar with automotive sales business and skilled at Excel.
  • Reporting and Dashboard Designing: Draft the final statistical portrait depicting the profit centers of dealers’ new car sales, financial and Insurance services, and maintenance. Indexed and visualized indicators for profitability with Tableau.

Accomplish­ments

Earners of the Tableau Desktop Specialist title use their knowledge of Tableau Desktop and data analytics to solve problems. They have demonstrated understanding of Tableau core concepts and terminology. Desktop Specialists are able to connect to, prepare, explore and analyze data, and share their insights.
See certificate
Foundational level knowledge of cloud services and how those services are provided with Microsoft Azure.
See certificate
Formulated informed the use of data integration platforms and how they relate to data pipelines and the ETL and ELT processes
See certificate

Projects

*
Oreoscoptes Distribution in BC
Introduction Wildlife preservation in British Columbia (BC) is an important topic to address. Having diverse wildlife could lead to a number of benefits, such as thriving ecosystems (e.g. diverse species lead to more balanced and resilient ecosystems; Ferrero & Troìa, 2021), or economic benefits (e.
Oreoscoptes Distribution in BC
Supervised Study on Heart Disease
Introduction Heart disease is a leading cause of death worldwide. Due to the potential for heart disease to cause negative effects, such as death, it may be important to investigate ways to prevent heart disease.
Supervised Study on Heart Disease
Watch out for homicides
Heroku: https://dash-app-murder.herokuapp.com Homicides in the United States is an important topic to think about. Compared to other high-income countries, the US contains greater homicide rates (e.g. gun homicides, https://pubmed.ncbi.nlm.nih.gov/30817955/). Because homicides can cause significant negative effects (e.
Watch out for homicides
Adaptive analysis of the heteroscedastic multivariate regression modeling
In this paper, we propose a novel method for the multivariate regression model, called multivariate weighted lasso (MWL). Numerical simulations and financial applications are conducted using the proposed and other existing methods. The results indicate that the proposed method performs well in both variable selection and coefficient estimation.
Adaptive analysis of the heteroscedastic multivariate regression modeling

Contact

Feel free to contact me if you are interested