Nearly 1.25 million new workers aged 15–29 are projected to join the workforce every month through 2022 in India. They will need to be skilled for the 21st century. Intelligent Learning Solution can create adaptable and affordable learning paths for learners, which can help them pick up new skills faster using a personalized learning path.
As a society our lives are getting transformed with Tsunami of Technology. This is creating a new kind of expectation to learn rapidly or become redundant. This is only the beginning. The world is changing fast and accelerating even faster. We all have to embrace a culture of continuous learning to remain relevant in the workforce. The challenge is that the new skills have steep learning curve and the learning content in many cases are beyond the affordability of masses.
Students are studying for jobs which do not exist yet!!
In Accenture Labs Tech4Good Program we explored this challenge. Our deep expertise in AI has helped create Intelligent Learning Solution that can create adaptable and affordable learning paths for Intelligent Learning Solution. When used in conjunction with affordable Open Source learning content, the learners can access a personalized learning path.
This intelligent Learning solution has been used in the context of NASSCOM Future Skills program that aims to rapidly and cost effectively reskill 1.5 to 2 million people in the country, and get jobs in the IT Industry. The solution effectiveness was demonstrated in a reputed engineering college.
It is a truly
ground-breaking platform as it consists of:
• Conversational user interface or bot: The bot, TutorBot, first solicits a learner’s intent and topic of interest, and
then offers guidance.
• Recommender engine: The bot invokes a recommendation engine in the backend to match the learner’s interest with relevant content. This recommender engine is a hybrid of content analysis and collaborative filtering of other learners’ interests
• Learning Path Generator: This component implements:
o Merging of Table of Contents (ToCs) of two authoritative freely available courses. Input can be YouTube playlists or any massive open online courses (MOOCs). The resulting table of contents is represented as concept nodes of the learning pathway.
o Depth of learning content composition that we call “Richness of Content” is arrived at by comparing the learning content composition with Wikipedia’s technical content.
o Latent Dirichlet Allocation based semantic similarity helps determine the hierarchy decision of the learning path
while merging two courses.
o Learner knowledge tracing uses the Deep Knowledge tracing approach.
o Ant Colony Optimization drives the Learning Path optimization.
•Knowledge assessor recommends appropriate content based on a learner’s assessment score.
•Learner progress tracker: The platform keeps a track of the learner’s progress on the completion, assessment scores and feedback of learning guidance.
With applicability in many similar learning opportunities, the solution will help students get ready for the Jobs Of the future!!