As part of supporting Taylor Violet Ross’s thesis, I assisted with the technical analysis by aggregating and processing data using Python. After generating similarity scores through AWS Rekognition and CompreFace, I helped consolidate the outputs into structured datasets, cleaning and organizing the results to ensure consistency. I used Python libraries such as Pandas, NumPy, and Matplotlib to visualize these similarity scores, creating charts that highlighted patterns and discrepancies between the different craniofacial approximation outputs. These visualizations helped Taylor clearly illustrate how different reconstruction approaches affected facial similarity, giving her a more data-driven foundation for her conclusions.
I also supported Taylor by reviewing and editing sections of her written thesis, ensuring that the technical findings were presented clearly and accurately. This included refining explanations of the analysis process, strengthening descriptions of the visualized results, and improving the overall clarity and flow of the research narrative. By helping integrate the visual data, similarity metrics, and methodological discussion, I contributed to making her final submission more coherent, accessible, and academically polished.
