top of page
Research & Projects
Random Forest Similarity Maps:
A Scalable Visual Representation for Global & Local Interpretation
In this Thesis, we present RFMap, an interactive visualization tool that supports explanation of RF models globally and locally by preserving the forest context. It uses dimensionality-reduction techniques such as UMAP & MDS to visualize the forest of trees & explain the structural relationships between the data instances from the model’s perspective.
Published in MDPI, Special Issue 10th Anniversary of Electronics: Recent Advances in Comp Science
bottom of page