illustration of a man in a suit pushing a rock up a hill.
1. Document upload
First, the builder uploads a document or file to your AI agent's library. The file can be a web page, a pdf or another supported format, which is part of the AI knowledge base.
2. Document conversion
- pdf, web pages, etc. - The system converts these files to a standardized text format, making it easier for AI to process them and retrieve relevant information.
3. Grouping and storage
The converted document is then divided into smaller, more manageable chunks. These chunks are stored in a database, allowing the AI agent to efficiently search and retrieve relevant sections during a query.
4. User inquiry
Once the knowledge bases are created, the oman mobile phone number user can ask the AI agent a question. The query is processed using natural language processing (pln) to understand what the user is asking.
5. Knowledge retrieval
The AI agent searches through the stored fragments, using retrieval algorithms to find the most relevant fragments of information from the uploaded documents that can answer the user's question.
6. Generation
Finally, the AI agent will generate a response by combining the retrieved information with its language model capabilities, crafting a coherent and contextually accurate response based on the query and retrieved data.
Advanced Gar Features
illustration of woman in suit standing on light bulb with telescope.
If you're not a developer, you might be surprised to know that not all rags are created equal.
Different systems will build different gar models, depending on their needs, use cases, or skills. Some AI platforms offer advanced gar features.
Semantic Chunking vs. Naive
Naive chunking involves dividing a document into fixed-size chunks, such as cutting text into 500-word sections, without regard to meaning or context.
Semantic fragmentation, on the other hand, divides the document into meaningful sections based on content. It takes into account natural breaks, such as paragraphs or topics, and ensures that each piece contains coherent information.
Mandatory citations
For industries that automate high-risk conversations with AI - such as finance or healthcare - appointments can help instill confidence in users when receiving information.
Developers can instruct their gar models to provide citations for any information submitted.
For example, if an employee asks a chatbot for information about healthcare benefits, the chatbot can respond and provide a link to the relevant employee benefits document.
Create your own AI agent with rag
Combine the power of the latest LLMS with your unique business insights.
Botpress is a flexible and infinitely extensible AI chatbot platform. It allows users to create any type of AI agent or chatbot for any use case, and offers the most advanced rag system on the market.
Integrate your chatbot into any platform or channel, or choose from our library of pre-built integrations. Start with the tutorials on the botpress YouTube channel or with the free botpress academy courses.
