– Organisations are generating unstructured data through various digital devices and platforms.
– BigQuery integrations with Vertex AI are being used to interpret unstructured data and extract meaningful insights.
– Multimodal generative AI capabilities are being used to analyze images and videos in BigQuery for various use cases.
Organizations are now generating unstructured data in the form of images, audio files, videos, and documents through digital devices and platforms like social media, mobile devices, and IoT sensors. BigQuery has introduced integrations with Vertex AI to make sense of this data by leveraging various AI tools like Gemini 1.0 Pro, PaLM, Vision AI, Speech AI, Doc AI, and Natural Language AI.
Vision AI provides image classification and object recognition capabilities, and now with multimodal generative AI use cases supported by Gemini 1.0 Pro Vision, organizations can analyze both images and videos using SQL statements directly within BigQuery. Capabilities like object recognition, info seeking, captioning, digital content understanding, and structured content generation are used in analyzing unstructured data in a data warehouse setting.
Gemini 1.0 Pro Vision can turn unstructured data like images into structured data, making it easier to analyze and combine with other datasets. By prompting Gemini 1.0 Pro Vision in BigQuery, organizations can access generative AI capabilities using the ML.GENERATE_TEXT() function, creating structured responses like HTML or JSON for deeper analysis.
These structured responses can be further refined in SQL queries to extract specific information like brand, model, and year from images of cars, enabling detailed analysis that can uncover valuable insights. Integrating generative AI directly into BigQuery offers significant benefits by enabling data analysis without the need for custom code or complex data pipelines. Organizations can unlock the potential of unstructured data and enhance their data workflows through this innovative approach.