Skip to main content

Zingg Community Icon

Zingg Community

Home
Events
Members

General
Help Zingg
Introduce Yourself
Zingg Databricks
Zingg Snowflake

⭐ Star Zingg
🐞 Submit an Issue
📚 Documentation

Powered by Tightknit

Understanding z_minScore and z_maxScore Metrics in Zingg Cluster Analysis | Zingg Community

Understanding z_minScore and z_maxScore Metrics in Zingg Cluster Analysis

·Apr 26, 2024 02:53 AM

Hi, can some explain the z_minScore and the z_maxScore associated with each cluster identified by Zingg? Thanks!

6 comments

· Sorted by Oldest

Sonal G.
·
Zaaheda I. these scores represent the least and most that record matches with another record in the cluster. this is mentioned in the docs as well.
Zaaheda I.
·
Hi Sonal - Thank you for your answer. I have read that bit online but I was hoping for a more detailed explanation like if the cluster has 5 records, the z_minScore value for one record in that cluster is the least it matches with which other record in the cluster? What does a higher value of z_minScore indicate vs a low score? I want to utilize these values to evaluate which matches to keep and which to reject and hence the need to know it in depth. Probably I am missing out on some documentations so feel free to drop a link if I am missing information. Thanks again!
Sonal G.
·
Happy to explain. Let’s say record a matches with b and b with c. A and c don’t match directly. Then an and c both will have min score 0. B min score will be min of its match with an and c.
🙌1
Zaaheda I.
·
Sounds good! What about the z_maxScore for these same records a, b and c? Can you explain that bit?
Sonal G.
·
A’s max score would be that of its match with B. Same for C. B would get the max of its match with A and C
✅1
Zaaheda I.
·
That makes sense. Thank you Sonal. 🙂
👍1