Introduction
Your Docebo learning platform is powered by Artificial Intelligence (AI), with several AI features included in your platform to enhance the way enterprises deploy and manage their learning programs, internally and across the extended enterprise, and is designed to deliver automated and personalized learning at scale.
All users in your platform - learners, managers, and administrators - can benefit from AI-powered features, and as with any features and functionalities in your platform, we want to ensure that you know exactly what’s going on and how to best use these innovative and time-saving features. This article outlines how to manage AI settings from the AI control panel. Learn more about AI in your platform.
Managing artificial intelligence settings
As a Superadmin, you can manage the settings related to the use of AI in your platform from the AI control panel. To access it, open the Admin menu by clicking on the gear icon in the top right corner of your platform, then press the Manage item in the Artificial intelligence section.
The AI control panel provides access to various features and settings that allow you to configure how AI is used in your platform. Let’s explore the available options and functionalities.
Auto tagging
You can use your platform’s AI to automatically generate tags to be associated with formal (training material) and informal content (assets), which will improve search features and suggestions. Moreover, it automatically assigns tags to the media files you upload to Creator’s media library. Learn more about auto tagging.
Use this area to enable or disable the auto tagging functionality and to manage its settings. On the auto tagging card, you can directly enable or disable the feature, or press Configure to access the auto tagging setting page.
At the top of the page, select the Enable auto tagging option to enable the feature. If you disable the auto tagging, the AI will stop automatically assigning tags to content, but the tags already assigned will remain.
In the AI retraining section, decide whether you want to take advantage of the retraining functionality. If you enable this option, the AI will learn from users’ tagging actions to improve the automatic association of tags to your content. We recommend enabling AI retraining for an optimal experience. If you disable the AI retraining option or the auto tagging feature altogether, any data that may have been collected for the improvement will not be lost. They will remain available to be used again if you choose to re-enable the AI retraining option. If you wish to request the deletion of the collected data, please contact Docebo using your platform's Help center.
In the Tag deny list section, you can insert a list of words that you don’t want to appear in automatically generated tags. When you save the deny list, the tags included in the list will be removed from all content already tagged in the platform, and they will not be used in future tags.
Remember that the deny list only works on tags auto-generated by artificial intelligence, not on tags manually added by users. AI already excludes by default any tags with words that are considered inappropriate (using the built-in profanity filter, see more detail in the next paragraph), then it excludes the tags included in the deny list.
While the profanity filter is a single list of words not allowed (offensive words, swear words, etc.) in all platforms, with the deny list you can define a specific list of words for your own platform, including words that you consider to be inappropriate for different reasons, such as competitors’ names and brands. Thus, words on the deny list are not necessarily offensive.
When you’re done with the configuration of the auto tagging feature, press Save changes. After you activate the auto tagging feature, all content that is subsequently added to the platform, or updated, will be automatically tagged. Content that was already present in the platform before the feature was enabled will not be auto tagged unless it is updated.
Notes about auto tagging
- To generate tags, Docebo’s AI requires a transcript of at least 30 words. If the transcript is shorter than this threshold, the auto tagging process is not triggered.
- Before generating tags, the system performs a quality check on the transcript. Tags are only generated if the transcript is clear and complete enough to be processed effectively.
- For security reasons, auto tagging is not performed on private content (that is, content requiring authentication to be accessed). When uploading private content, we suggest you add tags manually. Please note that only Superadmins and experts can add tags manually when uploading an informal asset, while users can add or edit manual tags only once the asset has been published.
- The auto tagging feature may encounter problems when processing SCORM and XAPI training material created through the use of certain authoring tools embedding textual contents directly into programming files, and compressing images so that their quality does not allow Optical Character Recognition (OCR). We are currently working to confront this limitation by collaborating with some of the major authoring tools providers.
- Any update to any training materials and assets triggers the auto-tagging feature to regenerate tags. These updates are logged as Training material has been updated events in the audit trail, performed by the user staff.support, in addition to the actual update event recorded under the platform user who made the change.
Best practices on audio and video for auto tagging
The auto tagging algorithm can also extract tags from any spoken audio of multimedia learning contents (both formal and informal) that you upload on your platform. However, there are some best practices for the correct interpretation of spoken audio by the state-of-the-art speech-recognition technology the company relies on:
- There are no significant background noises that could cover the spoken audio
- There is no or very little overlap between the speakers (if more speakers speak)
- The spoken audio language is the same throughout the recording
- A very strong speaker accent can lead to less precision
An audio track with any of the above characteristics extends the ability of speech recognition and comprehension technology to clearly distinguish spoken words and turn them into text, on which the automatic encoding algorithm operates. In these cases, the suggested tags produced through auto-tagging may be poor, irrelevant, or non-existent.
Auto skill assignment
If you are using skills, you can let your platform’s AI automatically associate skills to formal (courses) and informal content (assets), helping learners select the right content for their development. Learn more about auto skill assignment.
Use this area to enable or disable the auto skill assignment functionality and to manage its settings. Press Configure on the auto skill assignment card to access its setting page.
Please note! Make sure that you have completely configured the skills available in your platform in the Skills Management page before enabling the auto skill assignment feature. Once you enable the feature, the auto skill assignment process will begin and may take several hours to complete, depending on the amount of content to analyze. The automatic assignments cannot be reversed, and if you launch the process again, AI will only assign skills to content that was not previously assigned automatically, or to content that was updated after the automatic skill assignment.
At the top of the page, select the Enable auto skill assignment option to enable the feature. If you disable the auto skill assignment, AI will stop automatically assigning skills to content, but the skills already assigned will remain.
In the AI retraining section, decide whether you want to take advantage of the retraining functionality. If you enable this option, the AI will learn from users’ skill assignment actions to improve the automatic association of skills to your content. We recommend enabling AI retraining for an optimal experience. If you disable the AI retraining option or the auto skill assignment feature altogether, any data that may have been collected for the improvement will not be lost. They will remain available to be used again if you choose to re-enable the AI retraining option. If you wish to request the deletion of the collected data, please contact Docebo using your platform's Help center.
In the Content type section, decide whether you want AI to automatically apply skills to courses, assets, or both. If you change this setting, the new setting will only apply to future assignments, while the skills already assigned will remain. Note that the automatic assignment only works with published content.
Scroll down to the Automatic assignment criteria section. In the Trigger threshold field, you can set the minimum number of skills that content should have. If any piece of content has less than this threshold, AI will automatically assign skills. AI will assign any skills it finds suitable for the content, up to the amount you set in the Maximum skills assigned by AI field. For example, if you set the trigger threshold to 2 and the maximum skills to 3, AI will analyze all pieces of content with less than 2 skills assigned, and assign up to 3 skills to each of them. If you change these settings, the new settings will only apply to future assignments, while the skills already assigned will remain.
Finally, in the Tolerance threshold section, you can decide whether you want AI to assign a higher number of skills with a lower level of confidence. This means that the skills assigned may be slightly less accurate, but the assignment will be more thorough. If you change this setting, the new setting will only apply to future assignments, while the skills already assigned will remain.
When you’re done with the configuration of the auto skill assignment feature, press Save changes.
The process will run in the background until skills have been assigned to all the relevant content. It will then assign skills to all the content that is subsequently added to the platform or updated. You can download the log of all content analyzed by AI. The log becomes available when the feature is enabled, and it will remain available if it is disabled. It is downloaded as a CSV file and contains the following information:
- content ID
- content name
- content type (course or asset)
- date of automatic assignment of skills by AI
- skills added by AI. If no suitable skills were found, the “skills” column will be empty. If there are skills assigned by AI, these will be displayed in the same column and separated by a comma.
Best practices for auto skill assignment
- It can be useful to identify the content to which the AI could not assign any skills and assign skills to the content manually.
- Auto skill assignment relies on content titles and descriptions. For an optimal experience, ensure that your content titles are informative and descriptions are of high quality. High-quality descriptions provide comprehensive information about the content, using complete sentences to clearly convey the topic. They are descriptive and avoid the use of acronyms and abbreviations that may not be universally understood. For example, instead of "AI tools for learning. Lots of features to explore.", use "Artificial Intelligence Tools for Learning. Discover a range of powerful artificial intelligence tools designed to enhance your learning experience."
- Note that the AI may not analyze content if the description is too short, so it's essential to provide sufficiently detailed descriptions.
Content recommendation
You can use your platform’s AI to give personalized suggestions, which will improve the learner experience by offering the content that best suits each user. Learn more about content recommendation. Use the Content recommendation card to enable or disable the feature.
AI features for Creator
You can enable AI features to get the most out of your Creator tool. Use the AI features for Creator card to enable or disable them. When AI features are disabled, you can still create Creator lessons, but without enhancements such as AI-powered lesson generation, text revision, translation, video presenter, and automatic end-of-lesson assessment generation. Before enabling AI features, you must review and accept the related terms and conditions (opens in a new tab).