What is AI agent?
AI Classification agent uses pre-trained LLMs to classify products (Classification step) or extract data (Extraction step) based on attributes defined as source. Unlike trained AI models, AI Agent does not require additional training or user feedback loops to improve performance, enabling faster onboarding and a smoother user experience.
How AI Agent Works in SDM?
It uses its pre-trained natural language understanding to interpret labels, descriptions, and other sources from your input file.
No Training Required
The model performs accurately out-of-the-box, eliminating the need for training datasets or ongoing model updates.
Simplified Validation Process
- Up to configuration, you decide whether you need to automate the classification or not.
- Choosing automated mode means that the system will give you only the product rows to review that the AI wasn't able to classify automatically. The classified rows are visible from the “checked by AI” tab in SDM
- Choosing “to check”, means that 100% of the product rows need manual verification, even if AI has found a match for category or family. The classified rows are visible from the “to check” tab in the SDM.
More info on SDM configuration for AI Classification agent in the Admin guide and in from our API.
Key Benefits of SDM AI agent
- Reduced Time-to-Value: Save up to 75% of the effort previously required for training-related tasks.
- Wider Compatibility: Supports multi-language (e.g., Japanese, German, Arabic).
-
Improved Accuracy: Tested across diverse datasets to ensure robust performance in real-world scenarios.
[beta] Custom Prompts for better accuracy
You can now enhance the AI of the classification step by writing a custom prompt. This feature allows you to provide specific instructions to tailor AI behavior to your needs.
How to Use AI Classification Custom Prompts?
- Go to your project > Configuration > Project, and click the link to access the Configuration App.
- Select the relevant project and the classification step you want to customize.
- You'll find a text box where you can enter instructions for the AI to refine its classification process.
How to write an efficient prompt?
👉 We recommend taking our Akademy course to learn how to create clear and effective prompts suited to your context.
Changing the attributes used as a reference by the AI model
For both Extraction and classification, you can control and modify which attributes are used by the model to generate the predictions in the “sources” section right besides the custom prompt:
Not sure which field to choose?
Select fields that are well-maintained and rich in useful information (titles, descriptions, features, etc). The more detailed and cleaner the content, the better the AI will perform.