README.md

page_type	languages	products	name	description	azureDeploy
sample	csharp	azure azure-cognitive-search	Tokenizer sample skill for AI search	This custom skill extracts normalized non-stop words from a text using the ML.NET library.	https://raw.githubusercontent.com/Azure-Samples/azure-search-power-skills/main/Text/Tokenizer/azuredeploy.json

Tokenizer

This custom skill extracts normalized non-stop words from a text using the ML.NET library.

The language used for stop word removal can be optionally specified with the languageCode parameter using the ISO 639-1 code. Supported languages are:

• Arabic(ar)
• Czech (cs)
• Danish (da)
• Dutch (nl)
• English (en), is the default language used if none is specified.
• French (fr)
• German (de)
• Italian (it)
• Japanese (ja)
• Norwegian Bokmål (nb)
• Polish (pl)
• Portuguese (pt)
• Spanish (es)
• Swedish (sv)
• Russian (ru)

Requirements

This skills have no additional requirements than the ones described in the root README.md file.

Deployment

tokenizer

Sample Input:

{
    "values": [
        {
 "recordId": "record1",
            "data": { 
                "text": "ML.NET's RemoveDefaultStopWords API removes stop words from tHe text/string. It requires the text/string to be tokenized beforehand.",
                "languageCode": "en"
            }
        }
    ]
}

Sample Output:

{
    "values": [
        {
            "recordId": "record1",
            "data": {
                "words": [
                    "mlnets",
                    "removedefaultstopwords",
                    "api",
                    "removes",
                    "stop",
                    "words",
                    "textstring",
                    "requires",
                    "textstring",
                    "tokenized"
                ]
            },
            "errors": [],
            "warnings": []
        }
    ]
}

Sample Skillset Integration

In order to use this skill in a AI search pipeline, you'll need to add a skill definition to your skillset. Here's a sample skill definition for this example (inputs and outputs should be updated to reflect your particular scenario and skillset environment):

{
    "@odata.type": "#Microsoft.Skills.Custom.WebApiSkill",
    "description": "Tokenizer",
    "uri": "[AzureFunctionEndpointUrl]/api/tokenizer?code=[AzureFunctionDefaultHostKey]",
    "batchSize": 1,
    "context": "/document/content",
    "inputs": [
        {
            "name": "text",
            "source": "/document/content"
        },
        {
            "name": "languageCode",
            "source": "document/language"
        }
    ],
    "outputs": [
        {
            "name": "words",
            "targetName": "words"
        }
    ]
}