Learn how to create and implement custom skills in Azure AI Search to enrich data and ace the AI-102 certification exam. Perfect for aspiring Azure AI Engineers!
Table of Contents
Question
Xerigon Corporation plans to build a custom skill for their Azure AI Search service.
You create the following custom skill definition:
{
"name": " Xerigon-custom-skillset",
"description": "A custom skillset for enriching data",
"skills": [
{
"@odata.type": "#Microsoft.Skills.Text.EntityRecognitionSkill",
"categories": ["Person", "Location"],
"inputs": [
{
"name": "XerigonText",
"source": "/document/content"
}
],
"outputs": [
{
"name": "XerigonEntities",
"targetName": "recognizedXerigonEntities"
}
]
},
{
"@odata.type": "#Microsoft.Skills.Custom.WebApiSkill",
"uri": "https://your-custom-api-endpoint",
"httpMethod": "POST",
"inputs": [
{
"name": "Xerigontext",
"source": "/document/content"
}
],
"outputs": [
{
"name": "customOutput",
"targetName": "customEnrichedData"
}
]
}
]
}
For each of the following statements, select True if the statement is true. Otherwise, select False.
Statement:
- Xerigon-custom-skillset is available for indexing.
- customEnrichedData is a field in the search index.
- The skill executes once for each document in /document/content.
Answer
True
- The skill executes once for each document in /document/content.
False
- Xerigon-custom-skillset is available for indexing.
- customEnrichedData is a field in the search index.
Explanation
XerigonEntities, recognizedXerigonEntities, customOutput, and customEnrichedData are available for indexing, but Xerigon-custom-skillset is not. These entities are outputs that can be possible output field mappings in the indexer definition. The output field mappings specify what is absorbed in the search index. You would then have to specify the output field mappings to define what is actually in the indexer. For example, the following indexer has the specified elements based on the output of customEnrichedData:
"outputFieldMappings": [
{
"sourceFieldName": "/document/content",
"targetFieldName": "customEnrichedData",
"mappingFunction": null
}
],
There are no “outputFieldMappings” defined in the JSON code, so it is not possible to determine what fields are in the search index.
Xerigon-custom-skillset is not an output. Therefore, it will not be available for an output mapping.
The skill executes once for each document in “/document/content”. With the skill context of “/document/content”, the split skill executes once for the content. If you add an /*, for example “/document/content/*”, the skill will execute for each instance in the collection. If you want to access any enrichments for nodes added by a skill, you will have to specify the full path for the enrichment. If you want to use the text from the descriptions node as an input to another skill, specify it as “/document/content/descriptions/*”.
Microsoft Azure AI Engineer Associate AI-102 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Microsoft Azure AI Engineer Associate AI-102 exam and earn Microsoft Azure AI Engineer Associate AI-102 certification.