Skip to main content

Evaluation Metrics

Evaluation Metrics for NeuralSpace Transliteration Service

Let us learn about the two different evaluation metrics that we use to examine the performance of Transliteration models!

  1. Jaro Similarity

The Jaro distance is a measure of edit distance between two strings; its inverse, called the Jaro similarity

It is a measure of two strings' similarity i.e, The higher the value, the more similar the strings are.


The score is normalized such that 0% equates to no similarities and 100% is an exact match.

  1. Levenshtein distance:

Levenshtein distance is a metric for calculating the dissimilarity between 2 strings. It is the minimum number of single characters that one needs to edit in one string (including insertions, deletions and alterations) to change it into the other.


When comparing 2 words, lower Levenshtein distance means they are more similar.