A Customizable Recommendation Engine for LMS

A configurable recommendation engine has enabled Harbinger customer to cater a wide range of end-clients and deploy recommended learning plan solution



  • A UK based company in the field of blended learning wanted to expand their user base by providing a customizable application for generating a learning plan
  • This application was lacking feature of integration with different LMS
  • Older Adobe Flash based application had hardcoded contents and static UI resulting into complete dependency on software vendor for customizations


  • Need for generalized integration interface to support integration with various flavors of LMS
  • Need to have algorithmic approach to formulate reliable decision making process based on user interaction
  • Need to support desktop version of the application along with tracking of user activity
  • Need for administrative interface to create and support customizable contents

Harbinger Solution

  • Harbinger suggested automation of configuration of app, rules and contents of the app by providing web-based administrative interface
  • Designed decision tree based algorithms and implemented using JSON and Flex technologies to generate recommended learning plan according to user input
  • Utilized shared objects to retain user interaction in the desktop version of the application
  • PHP based email mechanism implemented to make learning plan available for future use
  • Provided configurable wizard based web-interface to provide details such as LMS APIs and parameters, text and graphics

Technologies & Tools

  • Adobe Flex
  • Action Script 3.0
  • PHP
  • JSON
  • HTML


  • With the help of authoring tool client's dependency on software vendor was eliminated
  • Customer was able to cater to various custom requirements for interactivity, LMS integrations and organization specific rules with the help of configurable tool
  • Increased and widened user base with the application's capability to integrate with various types of LMSs as well as with firms not having LMS capability
  • Accurate learning recommendations to the end users enhanced their learning experience