One cannot perform multiplication if one does not know addition. Concepts in Maths and Science intrinsically hook up to each other. The learner who comprehends one concept will figure out a consequent one easily and swiftly. Embibe’s primary goal is personalised learning by adapting to the knowledge state of the student. Capturing the knowledge state is accomplished by monitoring the students’ interaction with the platform at the concept level. These interactions range from watching videos, practicing questions, taking tests, and even viewing test feedback. \u2018Concept Mastery\u2019 is the process of modelling these interactions to find if a student has mastered a concept.<\/p>\n\n\n\n
Modelling \u2018Concept Mastery\u2019 is inherently complex as it requires modelling human understanding and how a human acquires knowledge. How a person receives knowledge is usually recorded through interactions. Also, the history of students’ interactions on various concepts is not sufficient. This insufficiency leads to inaccurate identification of the strengths and weaknesses of the student. Moreover, if just 1s and 0s were enough to determine how successful a person would be, no successful person would have failed in school. Thus, the system needs to analyse the knowledge in multiple dimensions.<\/p>\n\n\n\n
Learn: <\/strong> The first step towards mastering any concept is to understand it. Nothing beats visual learning in creating mental images of given ideas.Embibe’s \u2018Learn\u2019 consists of the world’s best 3D immersive content, making learning simple by visualising complicated concepts. The learning experience is built on a solid foundation of the industry’s most extensive Knowledge Graph of 74,000+ concepts and 2,03,000+ competencies. Embibe has integrated all its learning content with its pedagogy of Knowledge Graph of 74,000+ concepts. It ensures deep personalisation across grades, exams, and goals.<\/p>\n\n\n\n Practice: <\/strong>It takes practice to master anything. The same goes for \u2018Concept Mastery\u2019. Embibe’s \u2018Practice\u2019 feature consists of more than 10 lakh interactive question units packaged into chapters and topics of top-ranked 1,400+ books. An adaptive practice framework further strengthens ‘Practice’ by personalising practice paths for each student through deep knowledge tracing algorithms. Using solvers and templates, it dynamically generates personalised questions at run-time. A recommendation engine for learning intervention provides automatic help through videos and hints when a student is struggling with a concept or competency in a query. Constant feedback of attempt quality after every question categorised into \u2018Too Fast Correct\u2019, \u2018Perfect Attempt\u2019, \u2018Overtime Correct\u2019, \u2018Wasted Attempt\u2019, \u2018Incorrect Attempt\u2019, and \u2018Overtime Incorrect Attempt\u2019 keeps the learner informed and aware.<\/p>\n\n\n\n Test: <\/strong>Embibe provides tests that cater to every stage in the life cycle of a student\u2019s preparation. In addition, students also get detailed test feedback that identifies their academic and behavioural gaps. For example, Embibe’s AI identifies and categorises the topics covered in a test into ‘Chapters You Got Right’, ‘Chapters You Got Wrong’ and ‘Chapters You Did Not Attempt’. Students can also check their sincerity score and understand the conceptual, behavioural and time management issues they need to work on to improve. <\/p>\n\n\n\n At Embibe, \u2018Concept Mastery\u2019 sits at the core of the learning outcomes engine. The system utilises Embibe’s Knowledge Graph with over 74,000 connected concepts to determine a student’s \u2018Concept Mastery\u2019. The Knowledge Graph allows the system to identify the root concepts where the student lacks to improve their current \u2018Concept Mastery\u2019 for their goal. Additionally, Bloom’s Taxonomy is used to further add another dimension to learning to segregate knowledge into understanding and application. Further, questions are tagged with difficulty-level to adapt to the student’s knowledge level. Thus, we solve the problem of modelling \u2018Concept Mastery\u2019 using Knowledge Graph, Bloom’s Taxonomy, difficulty level, and latent variables as dimensions to the billions of interactions collected over the past eight years.<\/p>\n","protected":false},"featured_media":0,"template":"","meta":[],"yoast_head":"\n