Fig. 1
From: Using DenseFly algorithm for cell searching on massive scRNA-seq datasets
The result of the cell type identification experiment. The x-axis is the hash length while the y-axis is the average Cohen’s Kappa score given by the five-fold cross-validation. The performance of all algorithms improved as the hash length becomes larger. The figure indicates that DenseFly outperforms FlyHash and SimHash under all chosen hash length conditions, and DenseFly still reaches a high score even with a relatively shorter hash length. The experiment proves DenseFly’s feasibility on the scRNA-seq data similarity search