Totara Engage's recommendations engine allows users to find content from both Totara Engage (e.g. resources or playlists) and Totara Learn (e.g. courses).

You can enable or disable Recommendations in Quick-access menu > Configure features > Engage settings.

The recommendations engine will not work until you install and configure the engine. Find out how to do this in the developer documentation.

The recommendations shown to users will be limited in a number of important ways to ensure that they are relevant:

Find out more about how the recommendation engine's text processing works in the technical documentation.

Configuring the recommendations engine

You can configure how the recommendations engine works by navigating to Quick-access menu > Plugins > Machine learning settings > Recommendation engine.

SettingDescriptionNotes
Number of items-to-user recommendationsSelect the number of items-to-user recommendations generated by the recommendations engine.The default/recommended value is 5.
Number of items-to-item recommendationsSelect the number of items-to-item recommendations generated by the recommendations engine.The default/recommended value is 5.
Number of related itemsSelect the number of related items that will be displayed in the Related tab of the side panel for resources, playlists and workspaces.The default/recommended value is 5.
Recommendation algorithm

Select the algorithm used to determine recommended content. Choose from:

  • Full hybrid: Uses content data, user metadata (user profile and competencies), and user-content interaction data
  • Partial hybrid: Uses content metadata and user-content interaction data
  • Matrix factorisation: Uses user-content interaction data only

User metadata consists of: id in the database, language, city/town (free text), country, interests, aspiring position, positions, organisations, current competencies scale, badges, and profile description (free text).

Content metadata consists of: content type (e.g. workspace, course, article, micro learning article, or playlists), topics, and text description (free text).

Interactions data consists of: user id, content id, interaction value (0 or 1), and time of interaction.

The Full hybrid algorithm has the highest level of granularity, but takes the longest to process. Matrix factorisation has the lowest granularity but is the fastest to process. By default this setting is set to Partial hybrid, which is the most balanced in terms of granularity and processing time.
Time to analyse interactionsSet the time period (in weeks) from which user-item interaction data will be drawn.The default/recommended value is 16 weeks.
File path for python executableEnter the file path to the python executable which will run the recommendations engine.-
Processing threadsSelect the number of cores/threads that can be used by the recommendations library. This number should always be lower than the number of physical cores available.The default/recommended value is 2.
Data directoryEnter the path to the directory where all recommendations data will be stored.-

Remember to click Save changes when you have finished configuring these settings.

Related content

When viewing content in Totara Engage such as resources or workspaces, users will be able to access a side information panel showing additional details about the content.

Within this panel users can click the Related tab to see a list of similar content. The following content will be displayed in this tab for each type of content:

Recommendations blocks

There are two blocks which can be used to display recommendations, both of which can be added to various dashboards or the user profile.

Recommended for you

The Recommended for you block can be used to show a range of Totara Engage and Totara Learn content based on the Recommendations engine.

The Recommended for you block can display four types of recommendations:

To add this block to a dashboard follow these steps:

  1. Navigate to the dashboard you want to add the block to and click Blocks editing on.
  2. Click the add button () in the block area you want to use.
  3. Select Recommended for you from the list.
  4. Click the cog icon () and select Configure Recommended for you block.
  5. Under Custom block settings select whether to use the Tile or List display type and select the Number of recommendations to display in the block.
  6. Select the type of content you want to include using the Recommendation type setting.
  7. Choose whether or not you want to display likes and ratings for items in the block. Note that if you have selected Courses or Workspaces for the Recommendation type you will not be able to change this setting.
  8. Click Save changes.
  9. Once you have added the block and viewed content on the site the recently viewed content will be displayed.

Recently viewed

The Recently viewed block shows the user the Totara Engage and Totara Learn content they have recently viewed or visited. These items can be displayed in a list format or as cards/tiles.

The following content types are included:

To add this block to a dashboard follow these steps:

  1. Navigate to the dashboard you want to add the block to and click Blocks editing on.
  2. Click the add button () in the block area you want to use.
  3. Select Recently viewed from the list.
  4. Click the cog icon () and select Configure Recently viewed block.
  5. Under Custom block settings select whether to use the Tile or List display type and select the Number of items to display in the block.
  6. Choose whether or not you want to display likes, comments and ratings for items in the block.
  7. Click Save changes.
  8. Once you have added the block and viewed content on the site the recently viewed content will be displayed.

This block can be placed on users' dashboards to allow them to quickly return to content they have started, or previously discovered but didn't have a chance to start.


© Copyright 2020 Totara Learning Solutions. All rights reserved.