Jinyin Wang

This study compares KNN, logistic regression, and random forest for predicting cardiovascular disease using a Kaggle dataset, finding random forest to be the most accurate.

The paper evaluates precision loss in metabolomics mass spectrometry data compression and provides thresholds to minimize impact on downstream analysis.

This paper introduces entropy-based decoy generation methods to improve FDR estimation in metabolomics and presents a tool for accurate FDR-controlled analysis.

The paper introduces Aird-ComboComp, a compression framework for mass spectrometry data that improves compression ratios and decompression speed compared to existing methods.

Aird is a computation-oriented mass spectrometry data format that achieves higher compression and faster decoding compared to existing formats.