Professional software developer skilled in creating efficient, scalable code using languages such as Java and Python. Strong focus on team collaboration, adaptable to changing project requirements, and driven to deliver impactful results. Expertise includes debugging, problem-solving, and maintaining effective communication with cross-functional teams. Known for reliability and proactive approach to meeting development goals.
• Contributed to the development of a railway data management system, focusing on data collection, processing, and analysis modules to ensure efficient and accurate handling of railway-related data.
• Utilized Python and SQL for data cleaning and processing, optimizing storage and retrieval logic, which improved data processing speed by approximately 20%.
• Assisted in designing and implementing a data visualization dashboard, enabling the team to monitor and analyze railway operation data in real time, supporting informed decision-making.
• Collaborated with the team to apply machine learning algorithms for anomaly detection in railway system data, helping the company identify and address potential risks earlier, thereby enhancing system safety.
Malicious Email Detection via Machine Learning Approaches 05/2021-07/2021
https://doi.org/10.1117/12.2624886
Supervisor: Professor Nick Feamster, University of Chicago
• Compared 6 algorithms and decided to design the system based on Naïve Bayes to classify spam corpus, pre- detect, and filter spam
• Collected email data from the Text Retrieval Conference; programmed with Dictionary in Python to extract important information; pre-processed and cleaned data, applying Map function in Python to transform text data into numeric data and adopting an open-source jieba to split sentences into single words
• Conducted K-Fold cross-validation to verify the effectiveness of Naïve Bayes regarding the accuracy, recall score, F1 score, and time cost; concluded Naïve Bayes algorithm has the highest success rate of spam filtering.