https://images.ctfassets.net/atupprxuq58o/4j1WY2ftqtAQkqMOHBn7P1/2ecbb05eb97336449209145da86e2e2b/ai-generated-8620394.jpg

AI-supported search with RAG

Search across sources with semantic understanding

Modern websites aggregate content from a wide variety of sources for a wide variety of target groups. With AI-supported on-site search, you can ensure that your users find the content that is relevant to them - in natural language and summarized by AI, just like they are used to from ChatGPT, Gemini & Co.

Due to the wide availability of language models, users have become accustomed to formulating their search queries in natural language. They also expect AI-generated summaries instead of endless hit lists. But how do you get a language model for your DXP that reliably searches your own products and services without relying solely on the "knowledge" it has been trained to use?

Retrieval-Augmented Generation

Retrieval Augmented Generation (RAG) is an advanced artificial intelligence technique that enhances traditional language models (LLMs) by giving them access to your internal data sources. A RAG-supported language model searches for relevant information in your structured data sources, such as products, FAQs and instructions, before generating an answer in order to provide a precise and fact-based answer.

How does RAG work?

Retrieval: The user asks a question. The system does not search the entire Internet, but searches specifically in a defined knowledge database (e.g. company documents, PDFs, databases) for information that matches the question semantically.
Augmentation: The relevant text passages found are passed to the LLM of your choice together with the original question. The prompt is therefore "augmented".
generation (generation): The LLM generates a precise, fact-based answer based on this new context and can specify sources.

Advantages of RAG

Higher accuracy: Reduces "hallucinations" (invented facts) of the AI.
Actuality: Enables access to data that was created after the model was trained.
Data protection: RAG systems can access company data without using it to train the public model.
Transparency: Answers can be backed up with citations.

Andreas DeckersCTO at Intentive
Intentive has been working with search-centered websites for many years. We implement powerful search technologies and generative language models for your individual requirements and support you with the introduction.