Precision and Recall of the Matcher against a Labelled Sample
This lesson is part of the course
In this lesson
You score your matcher against 20 hand-labelled pairs: predicted matches at a threshold, true and false positives, precision, recall and F1.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Normalising Company Names: TRIM, LOWER, SUBSTITUTE and Legal Suffixes Locked
- Blocking Keys: First Letter + Postcode to Limit Comparisons Locked
- Exact vs Normalised Matching: How Many True Matches Do You Find? Locked
- Character Bigram Similarity: Jaccard and Dice with MID and SEQUENCE Locked
- Token Overlap for Company Names: Word Jaccard and the Overlap Coefficient Locked
- Combined Match Score: Weights and Two Thresholds (Match, Review, Non-Match) Locked
- Precision and Recall of the Matcher against a Labelled Sample Locked
- De-duplication: Cluster IDs, the Golden Record and the Score Distribution Locked