Babel Street Analytics API

https://analytics.babelstreet.com/rest/v1/name-similarity

https://raw.githubusercontent.com/rosette-api/curl-examples/develop/examples/name_similarity.curl

Name Similarity compares two entity names (Person, Location, or Organization) and returns a match score from 0 to 1. Although not required, we strongly recommend specifying the source language when known, for better accuracy. If you do not specify the source language, the name similarity algorithm will guess the language.

Based on the text of the two names, Name Similarity infers values for any of the fields you do not specify, and uses these values when it calculates the match score. In cases where the language may be difficult to infer accurately, such as a Japanese or Korean name in Han script, you may improve the accuracy of the match score by specifying the field.

Name Similarity utilizes override tables and other tools specific to entityType when matching. You can improve the accuracy of the match score by specify the entityType to ensure the correct overrides are applied.

Name similarity uses many parameters to calculate the similarity score for two names. The default values of the parameters are tuned to perform well on most datasets. You can modify match parameters to optimize similarity scores for your data and business case by adding parameters to the request.

{"parameters": 
  {"parameterName": value}
} 

{
  "name1": {
    "text": "Влади́мир Влади́мирович Пу́тин",
    "language": "rus",
    "entityType": "PERSON"
  },
  "name2": {
    "text": "Vladimir Putin",
    "language": "eng",
    "entityType": "PERSON"
  },
  "parameters": {
    "deletionScore": "0.2"
  }
}

{
  "score": 0
}

{
  "supportedLanguagePairs": [
    {
      "source": {
        "transliterationScheme": "string",
        "script": "string".
        "language": "string"
      },
    {
      "target": {
        "transliterationScheme": "string",
        "script": "string".
        "language": "string"
      },
    "licensed": true
    }
  ]
}