Open Expert Project.

Responsive image

An information retrieval and delivery system for deep knowledge domains and specific subject matter.

TECHNOLOGY AREA(S): Human Systems; Information Systems

OBJECTIVE: Companies are seeking ways to enable machine-facilitated peer-to-peer knowledge sharing among management and employees. Employees have locally learned knowledge that needs to be shared with each other. The knowledge takes the form of practical knowledge and expertise gained through experience. The we envision a set of software solutions that enable the capturing, curation, distribution and consuming of web-enabled, multi-media content. This project specificallyv intends to develop software and a supporting concept of operation for solutions that enable curation and distribution of knowledge of multi-media knowledge products, and build machine learning- and/or artificial intelligence-based agents that link knowledge products based upon natural language understanding and content analytics. Curation could entail two principle applications: one for informal peer-to-peer comments, corrections, and clarifications; and a second for formal, authoritative review of content prior to wider dissemination. Automation would assist in finding related multi-media content so that old and new content could be presented together and compared. The objective would be the development of a composable, semi-automated, scalable, structurally sound, mechanism capable of dynamically curating multi-media content based on formal rules, heuristics, and machine learning algorithms. The desired tools would also allow competent authorities the ability to properly vet, edit, and manage distribution of content created by warfighters with legacy, formally created, learning content in order to maximize the relevance, utility and timeliness of shared content. The desired solution would enable Commands to compose multi-step curation workflows and enable management through assigned roles of cognizant personnel through an intuitive user interface. The Dynamic Curation Engine would: enable the rapid curation and routing of video/graphic/audio/text materials to specific individuals based on role assignments; support automated and manual content tagging; display side-by-side comparison of content; and create/store curation meta-data efficiently for managing content in a federated cloud-based distribution system. DESCRIPTION: The current approaches for collecting and sharing knowledge among warfighters is inadequate. The traditional focus has relied exclusively on training content delivered either via schoolhouses or to Naval commands as formally prepared content. It is too slow and inflexible. New materials are formally vetted over extended timelines and often pre-loaded on units/command computers prior to getting underway with no ability to update content until the next deployment. Further, a great deal of practical knowledge and understanding is continuously developed by sailors and marines through experience and, if it is shared, is shared informally through ad hoc, 1-on-1 interactions. A new peer-to-peer approach to knowledge sharing is needed to share information at the deck-plate level and within the Naval information technology environment. Efforts are currently underway to facilitate warfighters to create multimedia-based content on-demand. However, the Navy will require software tools to systematically share and curate that content within the approved IT infrastructure. Of interest are tools that facilitate all aspects of curation at both a peer-to-peer level and with authoritative doctrine and training commands. The desired tools would facilitate semi-automated content inspection for both completeness, relationship among other content, potential security/releasability issues, enabling data meta-tagging of content to facilitate brokering to interested parties and flagging potential problems with linked content. The curation tool(s) would facilitate not just the modification and inspection of content, but enable the rapid discovery of similar existing content and reference materials that would inform the curation process and identify potentially conflicting or supporting content for adjudication. The curation would need to support consumer-level commenting and annotation across a community of users, as well as enable formal vetting from the Naval Training Enterprise (NTE) and doctrine commands when appropriate.
The desired solution(s) will:
• Be responsive to the demands of the Navy’s Sailor 2025 and the Navy Ready Relevant Learning initiatives.
• Consider requirements for the application of artificial intelligence/machine learning (AI/ML) techniques for related content inspection and discovery for a range of multi-media content.
• Support operations on intermittently connected networks in which cloud-based content is managed asynchronously via offline catalogs and partial caches.
• Enable Commands to compose curation workflows and roles to different categories of curator/user roles based on the content and organization of the Command.
• Be designed to work within the existing Naval information technology environment to the greatest degree possible, and to be accessible in a range of DoD security enclaves (specifically, usable within the constraints of the DoD information systems).
• Be designed to facilitate the automated curation and distribution of quality content with minimal time requirements from content curators.
• Enable workflows for managing (i.e., detecting and deconflicting) new content with existing content, including cases where the new content originates from a different organization.
• Enable authoritative curation of previously-approved content in order to ensure that information remains fresh and relevant to consumer demands.
• Enable algorithms that facilitate efficient curation and identify issues at the enterprise level across large content data sets.
• Enable curator-driven content valuation through utility and scope metrics, and support annotation of content (to include comments and clarifications), for use in informing authoritative content managers (e.g., the Naval training community).
• Support unit-level distribution within a local Navy computer network and support the management of release authorities.
• Ensure that tools are available on-demand to sailors both ashore and on-board ship thru both Navy workstations.
• Enable a broad range of content to be curated, to include video, imagery, audio, and text, with structure provided via automated and/or semi-automated mechanisms (e.g., wizards).
• Support both enterprise-wide and command-defined checklists, which enable customized routing workflows on the basis of content creator identification of the presence of topics that require alternative or augmented curation review.

REFERENCES: Hussein, A. T. T., Singh, S. K., Farouk, S., and Sohal, A. S. “Knowledge sharing enablers, processes and firm innovation quality.” Journal of Workplace Learning, 28(8), 10 October 2016, pp. 484-495. is external)
Groza, T., Tudorache, T., and Dumontier, M. “State of the art and open challenges in community-driven knowledge curation.” Journal of Biomedical Informatics, 46(1), 2013, pp. 1-4. is external)
Dong, X. L. and Srivastava, D. “Knowledge curation and knowledge fusion: challenges, models and applications.” Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data, May 2015, pp. 2063-2066. is external)
Kochhar, S., Mazzocchi, S. and Paritosh, P. “The anatomy of a large-scale human computation engine.” Proceedings of the ACM SJGKDD Workshop on Human Computation, July 2010, pp. 10-17. is external)
Kim, K. and Tse, E. “Search engine competition with a knowledge-sharing service.” Decision Support Systems, Volume 66, October 2014, pp.180-195. is external)
Zhuge, H. “A knowledge flow model for peer-to-peer team knowledge sharing and management.” Expert Systems with Applications, Volume 23, Issue 1, July 2002, pp. 23-30. is external)
Razmerita, L., Kirchner, K. and Nabeth, T. “Social media in organizations: leveraging personal and collective knowledge processes.” Journal of Organizational Computing and Electronic Commerce, 24(1), January 2014, pp. 74-93.
Chennamaneni, A., Teng, J. T. and Raja, M. K. “A unified model of knowledge sharing behaviours: theoretical development and empirical test.” Behaviour & Information Technology, 31(11), pp. 1097-1115.
Learn more