Dai Ma

  • Phone: +86 19864895373
  • Email: 1425529116@qq.com
  • WeChat: hjdch467327
  • Date of Birth: 2004-09-24
  • Research Interests: Text Analytics, Data Mining, Recommendation Systems
Profile

Self-Evaluation

Driven by a solid grounding in algorithms, data structures, and software engineering, I have independently led multiple deep learning and computer vision projects using C++, Python, and PyTorch, demonstrating expertise in model design, optimization, and performance analysis. My analytical, data-driven mindset enables me to decompose complex challenges and deliver robust solutions. I excel in collaborative environments through clear communication and thrive in independent research by rapidly mastering new technologies. Passionate about scholarly exploration, I am dedicated to advancing text analysis and data mining theory and applications, with a strong aspiration to pursue doctoral studies and contribute to cutting-edge academic research.

Education

Sun Yat-sen University — B.Eng. in Computer Science and TechnologySept. 2022 – Jun. 2026

  • Core Courses: Data Mining and Machine Learning (96), Optimization Theory (92), Artificial Intelligence (91), Operating Systems (93), Computer Organization (91), Data Structures and Algorithms (95)
  • Academic Performance: GPA 4.0/5.0 (Top 15 %, Rank: 44/321); average score of all courses above 90
  • Awards:
    • 2023 Third Prize, Guangdong Province, National Undergraduate Mathematics Competition
    • 2024 Bronze Prize, Sun Yat-sen University Programming Contest
    • 2024 Third Prize, University-level Scholarship, Sun Yat-sen University
    • 2025 Meritorious Winner, Mathematical Contest in Modeling (MCM/ICM)

Research Projects

Risk Control Driven Adversarial Spam Text Detection via Character Similarity Network

Project Leader · Apr. 2025 – Present

  • Designed end-to-end spam detection pipeline: multi-source dataset collection, HTML stripping, lowercasing, regex cleaning, jieba tokenization, stop-word removal, BOW/TF-IDF/n-gram features.
  • Developed a character similarity network with glyph & phonetic codes, precomputed similarity matrix for adversarial Chinese variants; learned robust embeddings via graph clustering; trained LR and SVM classifiers.
  • Conducted experiments on clean vs. adversarial data, significantly improved recall and F1; fine-tuned similarity thresholds.

Integrated Smart Home Device Control System

Project Leader · Apr. 2025 – Present

  • Built unified platform managing IoT devices (lighting, HVAC, security) via web/mobile.
  • Developed dynamic device discovery and lifecycle operations; scene creation, scheduler, event-driven scripts.
  • Designed middleware to unify MQTT, ZigBee, BLE, HTTP protocols with consistent backend API; modular Vue.js + Node.js/Express + MySQL + MQTT architecture.

Maximizing National Performance in the Olympic Games

Project Leader, COMAP MCM · Sept. 2024 – Dec. 2024

  • Built medal prediction system combining XGBoost, BP neural network, chi-square tests, multivariate regression. Achieved 90 % F1 for medal types.
  • Sensitivity analysis confirmed robustness (F1 variation < 5 %).
  • Forecasted 120 medals for the USA in 2028; 42 % chance for Nepal’s first track & field medal.

Image Classification: Optimization and Model Comparison

Project Leader · Sept. 2024 – Dec. 2024

  • Enhanced CIFAR-10 classification by comparing Softmax, MLP, and CNN in PyTorch; data augmentation & normalization.
  • Tuned SGD, momentum, Adam; Adam achieved 30 % faster convergence, 40 % lower loss volatility; CNN reached 70 % accuracy.

Skills & Proficiencies

  • Programming: C++, Python, SQL
  • ML & DL: scikit-learn, TensorFlow, PyTorch; algorithms (regression, classification, clustering, ensembles), architectures (CNNs, RNNs, Transformers)
  • Foundations: Operating Systems (Linux, Windows), Computer Networks, Database Systems