Implement Semantic Search in Oracle APEX Using AI Vector Search
Introduction
Oracle APEX now leverages the AI Vector Search feature introduced in Oracle Database 23ai to implement semantics-based similarity searches. This feature lets developers run deep learning models and create vector embeddings without leaving the database. AI Vector Search transforms data into high-dimensional vectors, enabling advanced semantic search functionalities in APEX applications. Unlike traditional keyword-based searches, this technology understands the meaning behind the data, providing more relevant results by conducting context-aware similarity searches.
Oracle Database 23ai supports all modern data types and workloads, including vectors, and incorporates AI and machine learning capabilities directly within the database. By generating and storing vectors—multidimensional representations of documents, images, videos, sound, and other types of unstructured data—developers can encode these objects as vectors and enable searches for semantic similarities using mathematical calculations. This powerful technology allows for combining similarity searches with searches on business data using simple SQL, making it accessible to anyone with a basic understanding of SQL.
For those running APEX on top of Oracle Database 23ai, the AI Vector Search capabilities are natively available at no additional cost. This means that APEX developers can seamlessly include these advanced search functionalities in their applications, enhancing the user experience with more accurate and context-aware results.
Demo
Prerequisites and setup
- Oracle Cloud account—sign-up page
- Pre-trained ONNX model—documentation
- Oracle Database 23ai—documentation