Which Azure Language Service Recognizes Company Product Names and Codenames?
You are building a chatbot. You need to ensure that the bot will recognize the names of your company’s products and codenames. The solution must minimize development effort. Which Azure Cognitive Service for Language service should you include in the solution?
Community Votes
100% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.
Community Insight
The question tests matching a business-specific extraction requirement to the right Azure AI Language capability, and the common trap is assuming entity linking can discover your private products and codenames when it only resolves entities already present in a public knowledge base.
A chatbot must recognize your company's product names and internal codenames with minimal development effort, which points to custom Named Entity Recognition (NER) in Azure AI Language. This page establishes why option C is the correct service choice and why entity linking, key phrase extraction, and custom text classification do not satisfy the requirement.
The most frequent wrong pick is entity linking (B), because learners assume a knowledge-base link automatically finds their products; in reality entity linking cannot recognize private product names or codenames that do not exist in public graphs, and it offers no way to train on your own labeled terms.
Community Discussion (7 comments)
Comments & Corrections
No comments yet — spotted an error or have a note? Share it below.
Expert Analysis
Why the Answer Is Correct
Azure AI Language custom Named Entity Recognition (NER) is the only option designed to extract user-defined entity types from text, which is exactly what mapping company products and codenames requires. You label a small set of example sentences in Language Studio, train the model, and the hosted endpoint returns those custom entities in the chatbot's input. Because Azure handles training, versioning, and hosting, the development effort stays close to data labeling rather than building a bespoke NLP pipeline, satisfying the "minimize development effort" constraint. It also supports adding new products and codenames later by retraining on additional labeled examples. This makes option C the direct fit for a chatbot that must understand organization-specific vocabulary.Why the Other Options Are Wrong
Custom text classification (A) assigns whole documents or sentences to your own labels, so it can tell you that a message is about "products" but cannot pull out the specific product or codename mentions a chatbot needs. Entity linking (B) maps detected mentions to entries in a public knowledge base such as Wikipedia, so internal codenames and many product names simply will not resolve, and no customization is possible. Key phrase extraction (D) is an untrained, prebuilt capability that surfaces statistically salient phrases; it has no concept of your custom entity categories and will not consistently return product names or codenames as structured entities. Only custom NER combines user-defined entity types with a managed, low-effort training workflow.Community Comment Notes
All learner votes and comments on this question converge on C, and GHill1982's reasoning that custom NER lets you "recognize specific terms relevant to your business, such as product names and codenames" captures the mechanism precisely. Murtuza and syupwsh add that NER identifies and categorizes specific names or terms in text and that customizing it tailors the model to company-specific entities, which is the same conclusion reached here. One caution worth noting: the documentation link GHill1982 attached points to a computer-vision shelf-analysis article, which is unrelated to Azure AI Language custom NER, so treat that URL as a mis-link rather than supporting evidence. Shorter comments such as HaraTadahisa's "I say this answer is C" and takaimomoGcup's "NER is right answer" add agreement but no new technical argument, so the reasoning above remains the decisive basis.Official Reference
Exam Strategy
On AI-102, first decide whether the requirement is entity extraction, document classification, or linking to public knowledge, then pick the matching Azure AI Language feature. If the entities are private, business-specific, and must be learned from your own examples, custom NER is almost always the intended answer, and "minimize development effort" signals a managed prebuilt/custom service rather than a hand-built model.
Frequently Asked Questions
Why isn't entity linking enough for company product names and codenames?
Entity linking only resolves mentions against a public knowledge base, so internal codenames and many private product names return no match. Custom NER instead learns those exact terms from your labeled examples.
Does custom NER require much development effort for a chatbot?
No. Azure AI Language handles model training, hosting, and the prediction endpoint; your main effort is labeling example sentences and retraining when products or codenames change.