A simple python entity-resolver sample for the CORA dataset Resolves entities of possible duplicates of publications by comparing authors and the title tag.
- 72125 duplicates in gold-standard
- 20748 retrieved duplicates
- True positives: 16672
- True negative: 271103
- False positives: 4076
- False negative: 55453
- Precision: 0.80
- Recall: 0.23
- F1 Score: 0.359