Microsoft Certified: Azure AI Engineer AssociateImplement natural language processing solutionsMedium
A legal firm needs to process thousands of legal documents to automatically extract specific clauses, contract dates, and party names. These entities are highly specialized and not typically recognized by pre-trained named entity recognition (NER) models. The firm wants to build a custom solution without extensive coding. Which Azure AI Language feature should they use?
- APersonal Identifiable Information (PII) Detection
- BKey Phrase Extraction
- CPre-built Named Entity Recognition (NER)
- DCustom Named Entity Recognition (NER)
Show answer & explanationAnswer & explanation
Correct answer: D. Custom Named Entity Recognition (NER)
Since the entities are highly specialized and not covered by pre-built models, the firm needs Custom Named Entity Recognition. This feature allows users to define and train models to extract domain-specific entities from text without requiring deep coding knowledge.
Why the other options are wrong
- A. PII Detection identifies personal data like names or addresses, not specialized legal entities.
- B. Key Phrase Extraction identifies general important phrases, not specific, categorized entities like 'contract dates'.
- C. Pre-built NER would not recognize specialized legal entities like specific clauses or contract dates.
Custom Named Entity Recognition (NER)
An Azure AI Language feature that allows users to define and train models to extract domain-specific entities from unstructured text. This is essential when pre-built NER models do not cover the specialized entities required by an application.
- Extracts user-defined, specialized entities.
- Requires labeling custom entities in training data.
- Crucial for domain-specific text analysis (e.g., legal, medical).
Memory trick: Custom NER finds specific legal gems.