jev match

Decide whether pairs of records describe the same thing: same, different, or unclear.

jev match (--pairs <ref> | --left <ref> --right <ref> | --dedupe <ref>) [--kind <text>] [options]

When to use

Dedupe a contact export, align your catalog with a supplier's, merge duplicate issues. One Score per pair has three levels that map to actions: leave unlinked, send to a person, merge. The decision is the most likely level, so there is no threshold to tune.

Options

Option	Meaning
-p, --pairs <ref>	JSON: [["a","b"], ...] or [{"left": ..., "right": ...}]. Items are strings or {id, text}.
--left <ref> --right <ref>	Two JSON item lists; every left item is compared with every right item
-d, --dedupe <ref>	One JSON item list; every pair within it is compared
-k, --kind <text>	What the records are, e.g. "customer contacts". Sharpens the question.
--fail-on <list>	Exit 2 when any pair is same, unclear, or different

Limit: 200 pairs per call, sent in groups of 50. For large sets, block first (by postcode, name prefix, category) and match within blocks.

Output

results[] with left, right (ids), left_text, right_text, decision, confidence, probabilities (different, unclear, same); summary counts.

unclear is a real outcome meaning a person should look, not a weak same.

Example

jev match --dedupe @contacts.json --kind "customer contacts"

Decision   Conf  Left    Right
---------  ----  ------  ------
same       0.91  c_1042  c_2210
unclear    0.48  c_1042  c_3187
different  0.97  c_2210  c_3187

1 same · 1 unclear · 1 different

# Only the pairs a human should check
jev match --dedupe @contacts.json --json | jq -r '.results[] | select(.decision=="unclear") | "\(.left) ~ \(.right)"'

Pattern: entity alignment cookbook.