Social Media Python

SHIQI (SARA) SUN

*******@***.*** 347-***-**** linkedin.com/in/shiqisun

SUMMARY

Projects experience in various areas, including full-stack and machine learning application development. Strong programming skills in Java, Python, C++, SQL, HTML, CSS, and JavaScript, and solid background in financial engineering, mathematics, and statistics. EDUCATION

NEW YORK UNIVERSITY, TANDON SCHOOL OF ENGINEERING Brooklyn, NY Master of Science in Financial Engineering (GPA: 3.8/4.0) Expected 05/21 WUHAN UNIVERSITY Wuhan, CN

Dual Degree: Bachelor of Economics in Finance; Bachelor of Science in Applied Mathematics (GPA: 3.6/4.0) 06/19 Coursework: Advanced Algorithms, OOD, Data Structure, Big Data, Deep Learning, Programming Languages (Java/C++/Python), Database, Web Development, Cloud Computing, Statistics, Probability SKILLS & CERTIFICATIONS

• Languages: Java, C++, Python, SQL/NoSQL, R, HTML/JavaScript/CSS, JSON, MATLAB, Scala

• Tools: Eclipse, Visual Studio, PyCharm, MySQL/MongoDB, Atom, Git/Github, Spark, React, RESTful, AWS, Google Colab, TensorFlow, Jira, Bootstrap, jQuery, EJS, Google Cloud

• Certification: Bloomberg BMC

PROJECTS

UBS Machine Learning Project: Personalized Stock Recommendation (Python, SQL)

• Designed stock recommendation system for UBS sustainable equity fund using collaborative filtering algorithm.

• Built MySQL database to store fund holdings; built machine learning model to predict matching score of stocks for each fund.

• Invented algorithm to evaluate ranking performance which can be applied on multiple queries dataset.

• Generated features using AutoML platforms (H2O/SparkBeyond) to improve precision of recommendation.

• Deployed stock recommendation system to Amazon Web Service for demonstration. Stock Price Movement Prediction (C++, SQL, JavaScript)

• Developed a full-stack service using C++ to analyze impact of quarterly earnings release on stock price movement.

• Retrieved historical data of 494 stocks in S&P 500 within time window selected by user.

• Built MySQL database to store data captured from Zacks and Yahoo Finance.

• Improved web crawler performance with multi-threading by reducing crawling time from 72 seconds to 6 seconds.

• Built aggregation pipeline to group data into 3 partitions based on earnings surprise; used bootstrapping to facilitate visualization.

• Created user interface that enables users to retrieve stock information and visualize price movement for each group. Web Development: To Do List App (HTML, CSS, JavaScript)

• Developed an interactive Web App for users to add, delete, update and store daily to-do list; deployed server side to Heroku.

• Built RESTful APIs with Node.js and Express to handle HTTP requests and responses.

• Built MongoDB database to store to-do list information and hosted it on MongoDB Atlas cloud service. Machine Learning and Natural Language Processing: Social Media Sentiment Analysis (Python)

• Retrieved 100,000 tweets with tweepy API and stored these tweets in text files.

• Designed algorithms to label tweets by sentiment using emoticons & Harvard Dictionary; validated labels with TextBlob API.

• Built machine learning models (NB/SVM/RF/CNN/RNN) to predict tweets sentiment using bag-of-words and word embeddings.

• Evaluated model performance via 10-fold cross-validation & confusion matrix; improved AUC of SVM, CNN and LSTM to 0.93. EXPERIENCE

Quantitative Developer Internship, Harvest Fund Management Co., Ltd. 07/17 - 08/17

• Harvest Fund Management is top 3 joint venture fund company in China with US$130 Billion AUM.

• Implemented Rolling PCA algorithm in R using monthly time series of 25 market indicators to compute systemic risk indicator.

• Built Markov-switching autoregressive model to calculate transition probability to high risk level.

• Identified foreign exchange market has highest systemic risk contribution; successfully predicted stock market drawdown in 2018. AWARDS & EXTRACURRICULAR ACTIVITIES

• Swimming, National Third Level Athlete

• US Interdisciplinary Contest in Modeling (ICM), Honorable Mention, 2016

Contact this candidate