🛡️ Who can access the knowledge base?
The management, organization, and improvement of the knowledge base is exclusively restricted to platform administrators.
Your knowledge base contains expertise specific to your company. Didask AI makes it accessible and easy to use for your employees, whenever they need it.
As an administrator, your role is to manage this knowledge base and ensure that it contains high-quality information for your employees.
The knowledge base is accessible through the administration tab, in the Didask AI section.
🔎 How does Didask AI use your knowledge base?
Once the learner's request has been assessed, Didask AI will search your database to find the sources that best match the learner's needs. It will select the most relevant ones and filter out those that seem less relevant.
Once the search is complete, Didask AI will develop a response to the learner's need and provide them with:
a reliability rating: indicating whether the sources found enabled it to answer the question assuredly;
the sources used to develop its response: allowing learners to verify the context of the reply they received.
If a piece of information is not explicitly present in your knowledge base, Didask AI will not be able to accurately respond to the learner's request. This is because we have programmed it to be as faithful as possible to your company's information.
For example, if the learner's question is “Where is our data hosted?” and your knowledge base only contains the name of your hosting provider, Didask AI will not extrapolate the probable location of this provider's servers based solely on this name. To answer this question, the information must be explicitly present in your knowledge base: “our data servers are in Ireland, and our application servers are in France.”
🎓 A pre-filled database with your training courses
If you use the Didask Studio, we automatically retrieve in real time the content of your existing training courses. This integration requires no action on your part and includes all the textual content of your projects.
This pre-built database allows your learners to immediately ask questions about any of their training courses, whether it's to:
Clarify a concept covered in the course;
Obtain additional examples tailored to their context;
Review specific points before an assessment;
Explore a topic that interests them.
To go further, you can enrich this database with your specific company content: internal procedures, operational guides, technical documentation, or even resources from your integrated collaborative tools.
🚀 How can you improve the performance of your knowledge base?
A high-performance knowledge base is one that has limited “competition” between sources of knowledge. For example, if a learner asks a question about a feature of your product, the following will compete:
the resource center article written about this feature;
the development team meeting minutes that mention a bug in this feature;
the internal newsletter that mentions the release of this feature;
etc.
Didask AI has several filtering mechanisms based on relevance to the learner's request. As an administrator, you can further improve its performance through several simple actions:
1️⃣ Give your documentation explicit titles:
✅ “Data servers: location and infrastructure,” “workshop - the fundamentals of cybersecurity,” “CR: Design mockup review for the HERA R&D project”
❌ “Tech doc,” “Workshop 14_Dupont,” “R&D and design sync”
2️⃣ Remove information that is redundant or conflicting with other knowledge in your database:
Example: in your ingestion plan, there are two documents written by different teams on the same subject that contradict each other. ➡️ Without your input, Didask AI will not be able to find the source of truth. In this case, we recommend deleting the version that is no longer up to date.
3️⃣ Avoid sources that have limited validity over time:
Meeting minutes on minor decisions in progress;
“Draft” documents or documents awaiting approval;
Information that changes frequently (prices, temporary contacts).
💡 To go further: Once your knowledge base is rolled out, find out how to continuously optimize it using usage data:
Keywords: Didask AI, coach, learning assistant, knowledge base
Any additional questions? Feel free to contact us at [email protected]. Our team is available to help and support you with your projects! 💬