A naukri.com initiative
Cloudblog
2d
290
Image Credit: Cloudblog
Introducing AI.GENERATE_TABLE: creating structured data from gen AI models in BigQuery
- BigQuery, connected with Vertex AI, integrates advanced AI models like Gemini 2.5 Pro/Flash to analyze unstructured data.
- Google's AI models can extract key information from various data formats and generate structured insights.
- AI.GENERATE_TABLE() in BigQuery converts unstructured data insights into structured tables based on prompts and schema.
- To analyze images, create a table for image data and connect to the Gemini 2.5 Flash model in BigQuery.
- Using AI.GENERATE_TABLE() with images, specify prompts for city recognition and output structured data format.
- The model identifies cities in images, providing names, states, history, and attractions, showcasing its capabilities.
- AI.GENERATE_TABLE() can also extract structured data from medical transcriptions with specified schema.
- For medical transcriptions, the model accurately extracts information like age, blood pressure, conditions, diagnosis, and medications.
- AI.GENERATE_TABLE() facilitates easy data transformation in BigQuery for efficient analysis and integration.
- For more details and feedback on these features, refer to the documentation or contact [email protected].
Read Full Article
17 Likes
For uninterrupted reading, download the app