PRODUCTION SCENARIO
A media monitor needs the company names and the cities mentioned in each news article, each returned with the category it belongs to.
Which Azure Language capability should it choose?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Named entity recognition
The prebuilt NER feature 'can identify and categorize entities in unstructured text such as people, places, organizations, and quantities', so companies and cities need no training at all. Custom NER exists to 'train the model to recognize specialized entities specific to your use case', which these are not, and it costs a labelling effort for nothing. Key phrase extraction returns topics with no category. PII detection targets sensitive values and redacts them.