Gooru leverages big data, AI, and research to develop Navigator technology.
Navigated Learning employs AI techniques to individualize the learning experience and make individual learning at scale possible. From a corpus of content in any discipline, Navigator can use the transcripts to compute the competency framework for the discipline with topic analysis and deep learning techniques. Navigator curates a catalog of learning activities to machine classify learning activities to competencies, and compute vectors such as relevance, engagement, and efficacy. Navigator uses AI techniques to understand each learner and to compute their preferences, context, and citizenship. Navigator then uses the science of learning with its curated catalog and real-time understanding of the learner to rank suggestions of learning activities. Navigator is generalizable across languages, disciplines, and learners.
Navigator goes beyond codifying concepts to structure learning as a progression space of competencies metric space of competencies with partial ordering of dependencies, called the Navigator Competency Framework. A polyline across domains of competencies for each subject represents a learner's proficiency is called their Skyline. Navigator Competency Framework includes a polyline algebra to compute measures such as route, reroute, and mean-time-to-learn. The Navigator Competency Framework details the competency with factors such as depth of knowledge, common struggles on a concept, and decay functions, embeds a variety of curated learning activities, captures each learner's profile, and relates to local norms. Gooru uses LDA and word2/doc2-vec embeddings to compute Navigator Competency Framework from a set of documents that are then reviewed and finalized by discipline experts.
With Navigated Learning, Navigator can measure all facets of learning and amplify education and skills training. To fully locate and characterize a learner, the many facets of learning need to be measured or calculated. Non-cognitive skills, such as grit and perseverance and soft skills, such as communication and collaboration are integral to provide a detailed and consistent representation of the learner across all of their facets and disciplines. This enables the suggestion of personalized pathways based on an increasingly complete understanding of the learner.
Big data is captured across distributed systems and offline activities. Navigator uses xAPI records aggregated across Learning Record Stores to gather the big-data. Machine learning approaches use the xAPI data streams to curate learning activities and locatelocating learners with increasing precision. As learners engage in learning activities, Navigator continuously improves the curation of content and location of the learner.
The search engine for learning is based on the structure of the learning space, the curated catalog of learning activities, and an understanding of the user (learner or instructor). Navigator uses all of these signals in query analysis and ranking to keep the search results pedagogically aware and personalized.
Navigator encodes Event, Condition, and Action (ECpA) models using Principles of Learning as the overarching guide to make suggestions. These models trigger a ranked list of suggested actions based on events, learner conditions, and principles of learning. By operationalizing the learning principles with big data, Navigator ensures the suggestions offered to learners and their instructors are backed by learning theories and science.
Songer, Nancy & Newstadt, Michelle & Lucchesi, Kathleen & Ram, Prasad. (2019). Navigated Learning: An Approach for Differentiated Classroom Instruction Built On Learning Science and Data Science Foundations. Human Behavior and Emerging Technologies. 10.1002/hbe2.169.
Chaitali Diwan, Srinath Srinivasa, and Prasad Ram. Automatic Generation of Coherent Learning Pathways for Open Educational Resources, In Proceedings of the Fourteenth European Conference on Technology Enhanced Learning (EC-TEL 2019), Springer LNCS, Delft, Netherlands, 16-19 September 2019 .
Aparna Lalingkar, Srinath Srinivasa, and Prasad Ram. (2019). Characterizing Technology-based Mediations for Navigated Learning, Advanced Computing and Communications, 3(2), ACCS Publication, pp. 33-47.
Praseeda, Srinath Srinivasa and Prasad Ram Validating the Myth of Average through Evidences In: The 12th International Conference on Educational Data Mining, Michel Desmarais, Collin F. Lynch, Agathe Merceron, & Roger Nkambou (eds.) 2019, pp. 631 – 634.
Chaitali Diwan, Srinath Srinivasa, and Prasad Ram. Computing Exposition Coherence of Learning Resources, In Proceedings of The 17th International Conference on Ontologies, Databases and Applications of Semantics (ODBASE 2018), Springer LNCS, Valletta, Malta, October 22-26, 2018.
Sharath Srivatsa, Srinath Srinivasa. Narrative Plot Comparison Based on a Bag-of-actors Document Model. In Proceedings of the 29th ACM Conference on Hypertext and Social Media (ACM HT’18), Baltimore, USA, ACM Press, July 2018.
Lalingkar. A., Srinivasa, S. and Ram, P. (2018). Deriving semantics of learning mediation, In Proceedings of the 18th IEEE International Conference on Advanced Learning Technologies (ICALT), IEEE, 9th July to 13th July, IIT Bombay.
Aditya Ramana Rachakonda, Srinath Srinivasa, Sumant Kulkarni, M S Srinivasan. A Generic Framework and Methodology for Extracting Semantics from Co-occurrences. Data & Knowledge Engineering, Elsevier, Volume 92, July 2014, Pages 39–59. DOI: 10.1016/j.datak.2014.06.002.
Sumant Kulkarni, Srinath Srinivasa, Tahir Dar. 2018. Syncretic Matching: Story Similarity Between Documents. In Proceedings of ACM IKDD Conference on Data Science and International Conference on Management of Data, Goa, India, Jan 2018 (CODS-COMAD 2018).