OneData uses the OneData Matching API to compare the similarity between two database objects containing a collection of key-value pairs. A match grader computes a single similarity score between the two objects being compared. The similarity score is a floating point number between 0.0 and 1.0, and is computed by comparing the key values. The similarity score computed depends on the similarity function configured for each individual key. The final match score for the two objects is determined by computing the weighted arithmetic mean of the individual key scores.

Consider the following example:

OneData computes the score for the keys company_name, street, and city. If m is the total number of fields in both objects, s i is the individual score of the ith key, and w i is the weight of the ith field. Then the final score is computed as: