Jaccard Index

Last revised by Candace Makeda Moore on 10 Sep 2020

The Jaccard Index or Jaccard similarity coefficient is a statistical tool which aids in the evaluation of the overlap in two sets of data. This tool is similar to the Dice coefficient but mathematically different and typically used for different applications. The equation for this tool is, where X and Y are sets, | the intersection of sets X and Y | / (| the union of sets X and Y|). This tool is well suited for the evaluation of object detection where the area detected can be understood as a set of pixels or voxels as well as several other tasks in AI. When this statistic is used in terms of AI for object detection it is sometimes reffered to as the IOU (intersection over union) metric.

ADVERTISEMENT: Supporters see fewer/no ads

Updating… Please wait.

 Unable to process the form. Check for errors and try again.

 Thank you for updating your details.