Weicheng Zeng Email & Phone Number

Virginia, United States

Built data pipelines, pricing intelligence for airlines in optimizing their pricing strategies and understanding market dynamics

New York, New York, United States

I have worked on three projects during my work here:- 4D MRI Brain Tumor Segmentation; - Multimodal Acute Pancreatitis Severity Prediction;- Hepatocellular Carcinoma Recurrence Prediction;

New York, United States

I have worked on two positions1. Conduct data-driven research on elder abuse and elderly welfare across various contexts and countries (Israel, Ukraine, Sudan) using YouTube, Reddit, and Twitter APIs with Dr. Tina Kilaberia2. Engineered a Website for Child Language Lab @ NYU, with Dr. Ailís Cournane

New York, United States

Sales data forecasting with machine learning models; full stack development.

New York, United States

- Utilized Python and OpenCV for frame extraction and annotation, and PyTorch for building a segmentation model to analyze antiferromagnetic image fluctuations.- Employed Python's Pandas and Statsmodels for in-depth statistical time series analysis, including autocorrelation and Fourier transform techniques, to identify and quantify trends and patterns in segmented frame data.- Leveraged PyTorch to train a predictive GAN model, enhancing the accuracy of next-frame predictions by 20% in the video sequence. - Engaged in active collaboration with domain experts, integrating data science insights with specialized knowledge to refine and validatethe analytical approach.

Santa Barbara

- Processed time series data using Pandas library in Python and conducted trend analysis of daily and seasonal energy generation using10 statistical modeling techniques- Explored risk patterns inherent in predictive models of 264 wind farm assets and 214 solar farm assets spanning a year in Texas byimplementing data visualization methodologies with Matplotlib- Leveraged standard deviation, quantile method, and kendall’s coefficient to identify high-risk wind and solar assets, delivering crucialinsights for reliability of the probabilistic model and potentially leading to higher energy efficiency

Santa Barbara, California, United States

- Collaborated with SLAC SSRL to develop Convolutional Neural Networks (CNN) models using deep learning techniques to automatethe analysis of X-ray diffraction images, expediting protein crystallography processes- Trained models on the campus SLURM cluster, developed multiple ResNet-based models to classify images as single or multi-latticewith a 94% testing accuracy on simulated data- Simulated a dataset consisting of 155,000+ images for model training, applied cross-validation and fine-tuning, and tested the modelon user data, achieving a <10% error rate- Contributed a report to Acta Crystallographic Section D to deploy a machine learning model at SLAC SSRL for broader application inX-ray research

Santa Barbara, California, United States

- Designed and executed psychological experiments with MATLAB's Psych Toolbox and Python, employing data analytics on 5,000+data points on 150 subjects to meticulously capture and analyze gaze patterns during face observations- Managed an end-to-end data pipeline, overseeing the collection of 500+ hours of eye-tracking data and preprocessing to integrate dataand optimize statistical robustness- Utilized the Statistics and Machine Learning Toolbox, processing 10,000+ data entries with Python and MATLAB, and crafted 10+ visualizations used for a research paper, supporting eye tracking and face recognition-

Guangzhou, Guangdong, China

- Researched and collected pharmaceutical data in China from multiple sources, integrated into a computable dataset.- Executed SQL querying, cleansed and imported trading data in JSON format into the MySQL Database- Utilized Python to integrate and store market data and identified critical customer trends and made time-series forecasts with SARIMA, LSTM and CNN techniques.

Master'S Degree, Data Science

Bachelor'S Degree, Applied Mathematics & Statistics And Data Science & Psychology