How to Improve Recall in Custom Named Entity Recognition
You are building a custom named entity recognition (NER) model using Azure AI Language. After labeling 200 documents, you train the model and achieve 85% precision but only 60% recall. Which action is most likely to improve recall?
Quick Answer
The answer is to increase the number of labeled documents, especially those containing the target entities. This works because recall measures the model’s ability to find all relevant instances of an entity, and low recall typically means the model is missing patterns that appear in your data. By adding more labeled examples that feature the target entities, you provide the custom NER model with richer, more varied contexts to learn from, directly addressing the root cause of missed predictions. On the Microsoft Azure AI Engineer Associate AI-102 exam, this scenario tests your understanding of how data quality and quantity affect model performance, often appearing as a trap where candidates mistakenly adjust confidence thresholds or training hours. Remember, precision is about being correct when you guess, but recall is about not missing any true positives. A helpful memory tip: recall is like a net—widen it with more labeled examples to catch every fish.
⚠ Common exam trap
Watch out — candidates often confuse confidence threshold tuning with a data quality fix, thinking lowering the threshold will magically fix recall, when in reality it only trades precision for recall without addressing the root cause of insufficient training examples.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Increase the number of labeled documents, especially those containing the target entities
Low recall in a custom NER model typically indicates that the model is failing to identify many instances of the target entities. Increasing the number of labeled documents, especially those containing the target entities, provides more positive examples for the model to learn from, directly improving its ability to recognize those entities and thus boosting recall.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Lower the confidence threshold
Why it's wrong here
Lowering threshold may improve recall but reduces precision.
- ✗
Increase the training hours
Why it's wrong here
Training hours are not adjustable by users.
- ✓
Increase the number of labeled documents, especially those containing the target entities
Why this is correct
More examples improve recall.
- ✗
Switch to a different Azure AI Language feature
Why it's wrong here
Stick with custom NER; other features are not designed for this.
Go deeper
Related to this question
About these practice questions
One of 945 original AI-102 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
Same concept, more angles
2 more ways this is tested on AI-102
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. You are building a chatbot that uses Azure AI Language to extract intents and entities from user utterances. The bot must recognize custom entities like product names that are not in the default model. Which feature should you use?
easy- A.Prebuilt entity recognition component.
- B.Key phrase extraction.
- ✓ C.Custom named entity recognition (NER) component.
- D.List entity in a conversational language understanding (CLU) project.
Why C: Custom named entity recognition (NER) is the correct feature because it allows you to train a model to identify domain-specific entities, such as product names, that are not included in Azure AI Language's prebuilt entity catalog. Unlike prebuilt components, custom NER uses a labeled dataset to learn the exact spans of text that represent your custom entities, enabling the chatbot to extract them accurately from user utterances.
Variation 2. You are building a custom entity extraction solution using Azure AI Language. You have a small dataset (50 documents) with annotated entities. You need to train a model that can extract similar entities from new documents. What is the best approach?
medium- ✓ A.Create a custom NER project in Azure AI Language and train it with your annotated data.
- B.Use the prebuilt entity recognition API to extract entities.
- C.Use the Conversational PII entity extraction feature.
- D.Use the built-in entity extraction skill in Azure AI Search.
Why A: Azure AI Language's custom NER (Named Entity Recognition) feature allows you to train a model using your own annotated dataset. With 50 documents, you have enough labeled data to fine-tune a custom entity extraction model that learns the specific entity types and patterns in your domain, enabling accurate extraction from new documents.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-102 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-102 exam.