Welcome to DivExplorer!
DivExplorer is a tool for analyzing datasets and finding subgroups of data where a classifier behaves differently
than on the overall data. To start, please upload a discretized dataset with both the output of a binary
classifier,
and the (binary) ground truth.
DivExplorer is brought to you by Eliana Pastor, Andrew Gavgavian, Elena Baralis,
Luca de Alfaro, and is described in
the papers:
-
Looking for Trouble: Analyzing
Classifier
Behavior via Pattern Divergence. Eliana Pastor,
Luca de Alfaro, Elena
Baralis. In Proceedings of the 2021 International Conference on Management of Data (SIGMOD '21), 1400–1412.
June
20–25, 2021, Virtual Event, China.
-
How Divergent Is Your Data?. Eliana Pastor, Andrew Gavgavian, Elena Baralis, and Luca de Alfaro. In
Proceedings
of
the 47th International Conference on Very Large Data Bases (VLDB), Demo Track, 2021. PVLDB, 14(12).