Data Analyst Machine Learning

Yue Fei

Phone: 949-***-**** Email: ******@***.*** Linkedin: www.linkedin.com/in/yuefei2000

Add.: 21Rockview, Irvine, CA, 92612

EDUCATION

University of California, Irvine (UCI) Irvine, US

Master in Electrical and Computer Engineering with concentration in Electrical Engineering 09/2023-06/2025 Huazhong University of Science and Technology (HUST) Wuhan, China Bachelor of Engineering in Electronic and Information Engineering 09/2018-06/2022 SKILLS

Languages: Python, C, SQL, JavaScript, HTML, CSS, Verilog Frameworks/Libraries: Pytorch, Tensorflow, Django, Caffe, numpy, pandas Tools: MySQL, Matlab, PyCharm, Code: Blocks, Keil5, Visual Studio Machine Learning: Logistic regression, random forest, SVM, linear regression, decision tree Deep Learning: CNN, image classification, RNN, transformer, LLM Data Analysis: A/B testing, hypothesis testing

INTERNSHIP EXPERIENCE

Microsoft Asia Internet Engineering Institute STCA Remote Data Analyst 07/2024-11/2024

Designed a detailed diagram of Microsoft’s subsidiaries, departments, and business areas using Figma FigJam.

Analyzed acquisition and merger records of major American companies since 1957 using Pandas and NumPy and created a business map of Microsoft with Matplotlib to highlight acquisition strategies and future business development.

Cleaned and processed a dataset of 3.37 million Spotify user reviews using Pandas and NumPy; applied NLTK for tokenization, stop word removal, and lemmatization to generate a word cloud.

Trained a Hugging Face Transformer model for sentiment analysis of user reviews, leveraging mixed precision and gradient accumulation techniques, achieving 72% accuracy.

Shanghai Enflame Intelligence Technologies Co., Ltd. Beijing, China Intern of Product Support Engineer, Product System Engineering Department (Beijing Outstation) 07/2021-08/2021

Gained practical experience with Linux commands, remotely managing testing machines, conducting system maintenance, and automating tasks using database management tools.

Assisted in testing the Velo platform, evaluating its remote monitoring capabilities and start/stop functionalities; authored test reports and helped visualize machine data through animated interfaces for platform enhancement. PROJECTS

Inferring Personal Attributes using GPT-4 Irvine, CA, US Independently 04/2024-06/2024

Utilized GPT-4 to infer gender and age from blog texts, achieving 63.6% accuracy for gender and an 8.03 MSE for age prediction.

Investigated the potential of LLMs like GPT-4 to extract sensitive personal information from publicly available data. Film data analysis and recommendation Irvine, CA, US Independently 11/2023-01/2024

Developed a recommendation system using Netflix data with collaborative filtering and matrix factorization algorithms in Python.

Leveraged an autoencoder-based deep learning model to extract user and movie features, achieving an accuracy of 0.547 and RMSE of 0.762 with a VAE model.

Predicting Moves of Chess Players Based on Deep Learning, HUST Wuhan, China Independently 12/2021-05/2022

Processed and segmented data using Python, including PGN file parsing, chess score transformation, and data mapping from player labels to six-channel matrix representations.

Developed a CNN with three convolutional layers, one fully connected layer, and a Softmax layer, using Caffe and PyTorch frameworks for move prediction, and trained the neural network model to predict chess moves with over 60% accuracy. Multimedia Technologies Capstone Project: Photo Search Questions WeChat Small Program Wuhan, China Team Leader 11/2021-01/2022

Implemented text recognition using the Pytesseract library and connected the front-end and back-end through the Django framework.

Developed the front-end of a WeChat mini program using HTML, CSS, and JavaScript on the WeChat Developer Platform Visualization of Seismic Data, HUST Wuhan, China

Research Assistant, Advisor: Associate Prof. Jing Xu 03/2021-05/2021

Developed a Python-based web scraper to collect data from earthquake monitoring sites.

Designed a dynamic web chart allowing users to interactively view earthquake distributions over time, with point size representing magnitude and detailed information displayed on hover.

Contact this candidate