We understand every student is different, with their unique learning style, pace, and preferences. That’s why our platform is designed to analyse your performance and provide targeted learning resources to help you improve your skills and knowledge. Embark on a personalised achievement journey with Artificial Intelligence in education to identify your strengths and weaknesses and learn more efficiently.

Personalised Achievement Journey (PAJ) aims to provide an optimal learning path to a student given their current knowledge level, a predefined time duration, a predefined curriculum, the importance of concepts for a student’s target exam and the effort required to master each concept. PAJ is of paramount importance for Embibe, an AI platform for delivering learning outcomes. At Embibe, we calibrate students’ concept mastery over contextualized knowledge graphs and behavioural profiles relevant to achieve learning outcomes. PAJ is an opportunity to expedite student learning by suggesting learning, practising and assessment packs in a hyper-personalised way.

The problem of designing an optimal learning path can be broken into two subproblems – selection of concepts and sequencing of learn objects, concept explainer videos and practice questions. We have modelled selection as a dynamic programming problem, which selects the most important concepts based on the time remaining to complete the Personalised Achievement Journey.

Selection of remaining to learn objects after each concept completion requires looking at value and cost aspects. Another consideration, along with selection, is to avoid excluding prerequisite concepts. And for this reason, we use subgraphs from our concept knowledge graph to evaluate the value and cost trade-offs instead of individual concepts.

Then it picks the subgroup concepts based on the value/cost ratio, where the value of the subgroup indicates the fundamentality and relevance in the final exam. In contrast, the cost is how much time the student requires to master that subgroup based on student-specific behavioural attributes and concept mastery on each concept, such that students can gain maximum learning outcomes.

The algorithm also decides whether the student requires learning content or not. It also provides the practice questions until users cover all the competencies under the given concept subgroup based on the adaptive practice algorithm.

Both of the above-mentioned conditions are based on a Markov Chain with its most recent state as concept mastery. It gets updated using a Bayesian Probability, also called Bayesian Knowledge Tracing. Moreover, the algorithm will be recalibrated at each step to ensure optimal time allocation.

The Personalised Achievement Journey algorithm also considers the prerequisite concepts from previous grades, which are important for the target exam. So, in case a student has poor fundamentals for a given syllabus, then the algorithm digs deep into the fundamental concepts and picks the bare minimum learning content from the previous grade, which helps students to fix their basics before advancing in the current grade.

References

[1] “#RAISE2020 – Embibe – AI-Powered learning outcomes platform for personalized education”, MyGov India, Oct 2020, https://www.youtube.com/watch?v=kuwFtHgN3qU

[2] Faldu, Keyur, Aditi Avasthi, and Achint Thomas. “Adaptive learning machine for score improvement and parts thereof.” U.S. Patent 10,854,099, issued December 1, 2020.

[3] Thomas, Achint, Keyur Faldu, and Aditi Avasthi. “System and method for personalized retrieval of academic content in a hierarchical manner.” U.S. Patent Application 16/740,223, filed October 1, 2020.

[4] Faldu, Keyur, Achint Thomas, and Aditi Avasthi. “System and method for recommending personalized content using contextualized knowledge base.” U.S. Patent Application 16/586,512, filed October 1, 2020.

[5] Faldu, Keyur, Achint Thomas, and Aditi Avasthi. “System and method for behavioral analysis and recommendations.” U.S. Patent Application 16/586,525, filed October 1, 2020.

[6] Desai, Nishit, Keyur Faldu, Achint Thomas, and Aditi Avasthi. “System and method for generating an assessment paper and measuring the quality thereof.” U.S. Patent Application 16/684,434, filed October 1, 2020.

[7] Dhavala, Soma, Chirag Bhatia, Joy Bose, Keyur Faldu, and Aditi Avasthi. “Auto Generation of Diagnostic Assessments and Their Quality Evaluation.” International Educational Data Mining Society (2020).

[8] Lalwani, Amar, and Sweety Agrawal. “What Does Time Tell? Tracing the Forgetting Curve Using Deep Knowledge Tracing.” In International Conference on Artificial Intelligence in Education, pp. 158-162. Springer, Cham, 2019.

[9] Agrawal, S., and A. Lalwani. “Analysing problem sequencing strategies based on revised Bloom’s taxonomy using deep knowledge tracing.” Proc ofInt ConfonIntelligent TutoringSystems (ITS). Berlin: Springer 407410 (2018).

← Back to AI home