Scientific discovery today depends as never before upon ease of access to data, associated sophisticated tools and applications, to enable research, education. Researchers who once worked in local, isolated laboratories now collaborate routinely and on a global scale.
Specialized instruments that werespread across multiple locations can now fit into a single lab connected via cyberinfrastructure resources and residing in big data. However, the sheer volume and heterogeneity of data bring a multitude of problems.
The primary intellectual merit of the proposed project comes from its vision of providing an augmented intelligence and cognitive support ecosystem (AICSE) assistants that enhances the capability of scholars and researcher in examining research topics, data and assists the user in understand its relevance to their goals. The expected result of this feasibility study will be a new resilient architecture for an agent-driven tool that is capable of ingesting any structured or unstructured data provided by the NASA and providing actionable insights. The project will produce a software design document or specification document that is composed of several layers that compass NASA-domain specific research areas . In particular the design will include, (1) a cross-platform user interface or agents, (2) natural language understanding functionality that maps tokens or words, sentences, paragraphs or documents to their respective meaning. As well as, produce document-specific sentiment, tone and intent information based on document corpuses. (3) Several open-source NLP models that can produce document-level or corpus themes, identifies relationships among themes and aggregate analysis. Also, the proof of concept, phase I effort will be limited to Google Scholar, because of its diverse research publication types, as well as the restricted timeframe of the project.
This feasibility study will examine agent-driven technologies from the standpoint of low-level cognitive tasks , thereby, reducing the cognitive workload of its end-user around concepts and ideas within the corpuses. The expected result of this feasibility study will be a new resilient architecture for an agent-based research assist tool that is capable of ingesting any structured or unstructured data provided by the NASA staff and providing actionable insights.
The work in this project will form the foundation for possible smart artificial agents designed to provide mentoring services in diverse commercial areas, in terms of connecting diverse to resources, such as (1)social service benefits, like SNAP and Medicaid, (2) employment assistance, (3) education/job training, (4) housing, (5) mental, behavioral/ physical services (6) substance abuse treatment