|
|
| A b s t r a c t |
| Title: | AI for Data: Vector Search in Oracle Database 23ai |
| Presenter: | Kai Yu |
| Organization: | Independent |
| Summary: | A key feature of Oracle Database 23ai is the introduction of vector search, which enables semantic content representation through vectors—arrays of numbers that encapsulate the meaning of data beyond mere words or pixels. This capability allows for advanced similarity searches based on semantic relevance. In this session, we will explore the AI vector search features of Oracle 23ai focusing on generating, storing, indexing, and searching vectors, and how these capabilities integrate with mission-critical enterprise functions. Particularly we will highlight three methods for generating vector embedding’s from unstructured data: 1.Pre-Created Embeddings: Loading vectors directly from external files. 2.Native Database Procedure: Using the VECTOR_EMBEDDING() function to import ONNX embedding models. 3.External Embedding Services: Integrating with services like OpenAI, Cohere, and Google. Additionally, we will demonstrate how to perform similarity searches using the VECTOR_DISTANCE function and discuss the performance enhancements offered by new vector indexes for faster and more accurate results. We will also examine how vector search enhances Generative AI and Large Language Models (LLMs), addressing issues like hallucination by enriching prompts with private database content. Using practical examples, we’ll show how to build a Retrieval-Augmented Generation (RAG) application utilizing Oracle’s AI vector search in conjunction with the LangChain framework and LLMs. Join us to discover how to leverage vector search for transformative insights and applications in AI! |
| Topic: | Analytics & Applications |
| Session Type: | One Hour Session |
| Target Audience: | |
| Experience Level: | |
| Session: | 8: 02/13/2025 9:35 am to 10:35 am |
| Location: | Room 100A |
| Biographical Sketch: | Kai currently is an independent Oracle technology consultant. Kai has worked for Dell Technologies as a Distinguished Engineer leading the solutions engineering for more than 22 years. Kai has 30 years’ experience of architecting various IT solutions by specializing in Oracle database, Oracle EBS, Oracle Cloud, Oracle BI Analytic. Since 2019, Kai has been focusing on AI, Machine Learning and Generative AI and LLMs Kai has been a frequent speaker at various IT/Oracle conferences such as IEEE HPCC, OOW, Collaborates, Oracle LAOUC tour, Oracle APAC tour, Oracle EMEA tours at RMOUG, HrOUG, SPOU,GPOUG, LUXOUG , DOAG(Germany), UKOUG, OUGN etc. Since 2019 Kai has given around 30 technical presentations related to Machine Learning and AI at various Oracle technology conferences in US and Europe and Asia. He also authored 36 articles in technical journals such as IEEE Transactions on Big Data . Kai has published two Apress books “Expert Oracle RAC12c” and more recent “Machine Learning for Oracle Database Professionals” Kai has been a Quest/IOUG conference committee member, IOUG RAC SIG president and IOUG Cloud SIG co-founder and the current vice president. Kai has been an Oracle ACE Director since 2010 and was featured as Oracle ACE Spotlight. He also received the 2011 OAUG Innovator of Year award the 2012 Oracle Excellence Award: Technologist of the Year: Cloud Architect by Oracle Magazine. |
|
|
| Conference Administration Tool | Copyright © 2025, Prairie Systems Group, Limited | |
| Page: DISPLAY_PRESENTER_ABSTRACT | Version 1.86 |