Data Analyst Analytics

Vu Hoang Phuong Anh

093******* — # ***********.****@*****.***

Summary — Final-year Computer Science student at Ho Chi Minh City Open University with an interest in data analytics. Familiar with Excel, SQL, and Python for data processing, analysis, and visualization. Experienced in academic projects involving research paper review, code reproduction, and solution improvement. Looking for opportunities to apply analytical skills in real-world business contexts as a Data Analyst intern Education

Ho Chi Minh City Open University (2022 - Present)

Bachelor of Science in Computer Science

– Current GPA: 2.88 / 4.0 (equivalent to 7.41 / 10.0)

– Achieved A+ in Probability & Statistics, Discrete Mathematics; A in Calculus

– IELTS 5.5 (reading: 6.5)

Skills

Programming Python, SQL( SQL server) Libraries Pandas, Numpy, scikit-learn, matplotlib Projects

Product Review Sentiment Analysis for E-commerce (Tiki.vn) March 2025 – May 2025

– Designed and implemented an end-to-end data analytics pipeline for product reviews collected from Tiki.vn, using Python and web scraping techniques.

– Cleaned and preprocessed large-scale Vietnamese textual data, including normalization, deduplication, and feature engineering for structured and unstructured fields.

– Built and trained a multi-class Logistic Regression model to classify reviews into positive, neutral, and negative sentiment groups; evaluated model performance with standard metrics and feature importance analysis.

– Developed a Flask web application that enables users to input a product link, automatically scrape recent reviews, predict sentiment distribution, and generate actionable purchase recommendations in real-time.

– Technologies: Python, Pandas, scikit-learn, Flask, NLP for Vietnamese, web scraping (Requests), joblib.

– Github: github.com/phganh0103/ProductReviewSentimentAnalysis Chess Game with AI Opponent June 2025 – August 2025

– Designed and implemented a complete chess game in Python with a user-friendly graphical interface using Pygame.

– Developed an AI opponent utilizing the NegaMax algorithm with alpha-beta pruning, move ordering, piece-square tables, and advanced evaluation heuristics (including hanging piece detection, repetition, and endgame conditions).

– Supported multiple game modes: Human vs AI (choose color), Human vs Human, and AI vs AI; provided features such as move log, undo, reset, and real-time board flipping.

– Engineered efficient state management, including move legality checking, castling, en passant, pawn promotion, and accurate endgame detection (checkmate, stalemate, threefold repetition, 50-move rule, and insufficient material).

– Technologies: Python, Pygame, multiprocessing for AI move calculation.

– Github: github.com/phganh0103/chess

Graph Transformer Model Enhancement for Fraud Detection July 2025 – Present

– Researched and engineered improvements to a state-of-the-art Graph Transformer Attention Network (GTAN) for financial fraud detection on tabular transaction networks.

– Enhanced the TransformerConv layer to incorporate edge features into the attention mechanism and gating, enabling richer relational modeling and better information flow between nodes.

– Tuned model hyperparameters (learning rate, max epochs, early stopping) for optimal convergence and generalization.

– Achieved significant improvements over the original baseline on multiple datasets:

– YelpChi: AUC 3-4%, F1 3-9%, AP 6-11%

– Amazon: AUC 1%, F1 (maintained), AP 1-2%

– S-FFSD: AUC 1-2%, F1 1%, AP 1-2%

– Technologies: Python, PyTorch, DGL, scikit-learn, YAML.

Contact this candidate