Yanshan Song

buyssong@bu.edu

I am currently a senior software engineer in Cloud Container Team of System R&D Department of Trip.com Group. My work focuses on the construction of container management platform based on Docker & Kubernetes. Before that, I got my MSc degree in Computer Science from Boston University in January 2021 and BEng degree in Computer Science and Technology in Soochow University(Suzhou) in June 2019.

Education

Boston University

Master of Science in Computer Science
September 2019 - January 2021

Soochow University

Bachelor of Engineering in Computer Science and Technology
September 2015 - June 2019

Experience

Senior Software Engineer

Trip.com Group

  • Cloud Native Application development based on Docker & Kubernetes
  • GPU virtualization and construction of Ctrip GPU cloud platform
  • CICD workflow design based on GitOps

May 2021 - Present

Python Web Scraping Engineer Intern

Suzhou Enhance Co., Ltd.

  • Used Beautiful Soup and Regular Expression to parse HTML and extract target dat
  • Realized a function of downloading Excel, Word, PDF files in web pages and extract useful data from these files
  • Cleaned the extracted data and stored the processed data in MySQL
  • Utilized Selenium to crack sliding verification code and other simple human-machine detection
  • Implemented over 70 web scraping programs and designed another program to dispatch and monitor these web scraping programs

May 2021 - Present

Research

Left and Right Operating Hand Recognition Detection System

Bachelor Degree Thesis

  • Developed a system based on Android that can recognize the current operating hand by analyzing the user’s sliding screen trace
  • Collected data of sample points in several known classified sliding traces with Android’s event response mechanism and calculated feature vector for each trace from length, velocity, displacement, shape four aspects.
  • Used Weka to train and evaluate models under different classification algorithms
  • Used the model with the best performance to predict the class of newly generated sliding traces

September 2015 - June 2019
Soochow University

Project

Web Caching

Network Course Project

  • Used Apache Traffic Server to build a proxy cache server between the server and the client that are 5000km apart, which increases the download speed of the client by 27.95%
  • Explored the role of keywords max-age and expires in terms of controlling cache expiration
  • Verified the eviction policy of the proxy cache of Apache Traffic Server

Novermber 2020 - December 2020
Boston University

Movie Score Prediction

Data Science Course Middleterm & Kaggle

  • Trained a movie score prediction model based on 16w+ movie review records with RMSE of 0.951 in the test set.
  • Used TF-IDF to vectorize movie comments and reduced data dimension by PCA
  • Utilized KNN to train the model and found the optimal value of k by analyzing F1 & RMSE of different models

April 2020 - May 2020
Boston University

Social Networks and Recommendation Systems

Data Science Course Project

  • The project is aimed at recommending new collaborations to researchers
  • Used Python library NetworkX to create social relationship graph for researchers
  • Used common friends method, Jaccard’s Index, Adamic/Adar Index to make recommendations

Boston University
April 2020 - May 2020

Recipe Recommendation App

Mobile Application Course Project

  • Developed an Android mobile application which could recommend recipes based on user selected ingredients
  • Developed a server-side program based on Spring Boot and deploy it on AWS
  • Used OkHttp framework to connect the client and server, and use fastjson to parse the json data returned by the server

Boston University
April 2020 - May 2020

Triangle Counting

Database Course Project

  • Implemented naive parallelization algorithm in Hadoop to count triangles
  • Implemented improved parallelization algorithm in Hadoop to count triangles

Boston University
May 2020