Extract Skills🔗︎

HTTP Verb	Path
POST	/v10/skills/extract

Extracts known skills from the given text.

Info

You can try this endpoint out at our Swagger page ( US Data Center | EU Data Center | AU Data Center )

Scope of use🔗︎

The purpose of this endpoint is to enrich data records using our Taxonomy. You may store the parts of the response that are relevant to individual data records - such as candidate or employee profiles - within these records. It is not permitted to cache, store, or utilize responses from this endpoint for the purpose of limiting request volume, or to create replicas or derivatives of the Service and the Taxonomy.

Request Body🔗︎

Text 🔗︎ string required

Text🔗︎

The text to extract skills from. There is a 24,000 character limit.

Language 🔗︎ string required

Language🔗︎

The language of the input text. Must be one of the supported ISO codes.

OutputLanguage 🔗︎ string

OutputLanguage🔗︎

The language to use for the output skill descriptions. If not provided, defaults to the input language. Must be one of the supported ISO codes.

Threshold 🔗︎ float

Threshold🔗︎

A value from [0 - 1] for the minimum confidence threshold for extracted skills. Lower values will return more skills, but also increase the likelihood of ambiguity-related errors. The recommended and default value is 0.5.

Sample JSON

{
  "Text": "",
  "Language": "",
  "OutputLanguage": "",
  "Threshold": 0.5
}

Response Body🔗︎

Info 🔗︎ object

Info🔗︎

Information explaining the outcome of the transaction.

Info properties

Code 🔗︎ string

Code🔗︎

Code	Description
`Success`	Successful transaction
`MissingParameter`	A required parameter wasn't provided
`InvalidParameter`	A parameter was incorrectly specified
`AuthenticationError`	An error occurred with the credentials provided

Message 🔗︎ string

Message🔗︎

This message further describes the code providing additional detail.

TransactionId 🔗︎ string

TransactionId🔗︎

The (GUID) id for a specific API transaction. Use this when contacting support@textkernel.com about issues.

EngineVersion 🔗︎ string

EngineVersion🔗︎

The version of the parsing/matching engine running under-the-hood.

ApiVersion 🔗︎ string

ApiVersion🔗︎

The version of the API.

TotalElapsedMilliseconds 🔗︎ integer

TotalElapsedMilliseconds🔗︎

How long the transaction took on Textkernel's server, in milliseconds. If the transaction takes longer to complete on the client side, that extra duration is solely network latency.

TransactionCost 🔗︎ decimal

TransactionCost🔗︎

How many credits the transaction costs.How many credits the transaction costs.

CustomerDetails 🔗︎ object

CustomerDetails🔗︎

Information about the customer who made the API call.

CustomerDetails properties

AccountId 🔗︎ string

AccountId🔗︎

The AccountId for the account.

Name 🔗︎ string

Name🔗︎

The customer name on the account.

IPAddress 🔗︎ string

IPAddress🔗︎

The client IP Address where the API call originated.

Region 🔗︎ string

Region🔗︎

The region for the account, also known as the 'Data Center'.

CreditsRemaining 🔗︎ decimal

CreditsRemaining🔗︎

The number of credits remaining to be used by the account.

CreditsUsed 🔗︎ decimal

CreditsUsed🔗︎

The number of credits used by the account.

MaximumConcurrentRequests 🔗︎ integer

MaximumConcurrentRequests🔗︎

The number of requests that can be made at one time. If using sub-accounts, this is the maximum number of concurent requests across all accounts, not just this single sub-account.

ExpirationDate 🔗︎ date

ExpirationDate🔗︎

The date that the current credits expire.

Value 🔗︎ object

Value🔗︎

Contains response data for the transaction.

Value properties

Truncated 🔗︎ boolean

Truncated🔗︎

Whether the input text was truncated or not due to length.

Skills 🔗︎ object[]

Skills🔗︎

A list of extracted skills.

Skills properties

Type 🔗︎ string

Type🔗︎

Type of skill. Possible values are Professional, IT, Language, or Soft.

Id 🔗︎ string

Id🔗︎

The ID for the skill in the skills taxonomy.

Confidence 🔗︎ float

Confidence🔗︎

A value from [0 - 1] indicating the overall confidence that the extracted term actually refers to a skill in the context of the text (the average value of the 'likelihood' values of the individual match scores).

Description 🔗︎ string

Description🔗︎

The description of the normalized skill concept in the requested language.

IsoCode 🔗︎ string

IsoCode🔗︎

The language ISO 639-1 code. This will only appear for language skills (Type = Language).

Matches 🔗︎ object[]

Matches🔗︎

A list of matches where this skill was found in the text.

Matches properties

BeginSpan 🔗︎ int

BeginSpan🔗︎

The index of the first character of the match (0-based)

EndSpan 🔗︎ int

EndSpan🔗︎

The index of the last character of the match (0-based).

Likelihood 🔗︎ float

Likelihood🔗︎

Likelihood that the matched term actually refers to a skill in the context of the text.

RawText 🔗︎ string

RawText🔗︎

The actual term that was found as evidence of this skill (the substring from BeginSpan to EndSpan).

Sample JSON

{
  "Info": {
    "Code": "string",
    "Message": "string",
    "TransactionId": "string",
    "EngineVersion": "string",
    "ApiVersion": "string",
    "TotalElapsedMilliseconds": 0,
    "TransactionCost": 0,
    "CustomerDetails": {
      "AccountId": "string",
      "Name": "string",
      "IPAddress": "string",
      "Region": "string",
      "CreditsRemaining": 0,
      "CreditsUsed": 0,
      "ExpirationDate": "2021-12-31",
      "MaximumConcurrentRequests": 0
    }
  },
  "Value": {
    "Truncated": false,
    "Skills": [
      {
        "Type": "string",
        "Id": "string",
        "Confidence": 1.0,
        "Description": "string",
        "IsoCode": "string",
        "Matches": [
          {
            "BeginSpan": 0,
            "EndSpan": 1,
            "Likelihood": 1.0,
            "RawText": "string"
          }
        ]
      }
    ]
  }
}