Google Cloud Natural Language API Tutorial: Get Started in 5 Minutes [2026]

What it does:

Identifies named entities such as people, organizations, locations, events, consumer goods, and other proper nouns within a text, returning their type, salience score relative to the document, and where applicable, links to Wikipedia metadata for entity grounding.

Use case:

What it does:

Returns a numeric sentiment score from -1.0 (negative) to 1.0 (positive) along with a magnitude value indicating overall emotional intensity, supporting both document-level and sentence-level analysis.

Use case:

What it does:

Combines entity recognition with sentiment scoring at the entity level, attributing positive or negative polarity to specific entities mentioned in the text — useful for understanding opinions about individual products or features within a longer review.

Use case:

What it does:

Tokenizes text and labels each token with part-of-speech tags, lemmas, dependency parse trees, and morphological features, providing the linguistic substrate for advanced NLP applications such as relation extraction or grammar correction.

Use case:

What it does:

Automatically categorizes documents into a hierarchical taxonomy of more than 700 categories spanning news, business, lifestyle, technology, and other domains, enabling automated tagging and topic discovery at scale.

Use case:

What it does:

When pre-trained models do not match a domain, AutoML Natural Language and Vertex AI let users train custom entity extraction, sentiment, and classification models from labeled examples and serve them through a managed endpoint.

Use case: