emi® is an AI-empowered decision-support platform for enterprises to draw collective intelligence to optimize business decisions.

It is a proposed spinout from the GoPeaks research lab at UNC Charlotte. The emi® system is a digital platform that aggregates collective intelligence from domain experts into knowledge graph-based, prescriptive analytics solutions for enterprise performance management (EPM).

Managers and executives are unable to optimize organization performance decisions because existing enterprise knowledge is fragmented and often contained within silos. This is a complex, widespread problem that adversely affects performance in organizations of any substantial size in most industries. This was confirmed through National Science Foundation I-Corps customer discovery interviews with 150+ senior practitioners in enterprise performance management (EPM). 

Initial opportunities have been discovered for performance management in healthcare systems and in financial institutions. Financial institutions, for instance, must meet quarterly financial return targets while funding expenditures on innovative services to retain customers and deliver employee satisfaction. The International Federation of Accountants estimates that fragmented reporting systems may cost the financial industry $780 billion annually. Healthcare systems need to improve patient care quality and outcomes, enhance clinician satisfaction, and increase resource use efficiency while meeting investor or other stakeholder requirements. A 2020 Gartner survey found that more than 50% of healthcare systems worldwide are not satisfied with EPM analytics software, quoting that the most critical impediment for long-range planning is fragmented insights.

Specifically, the system is built on several proprietary inventions. First, the system has the world's first integrated knowledge graph of causes-and-effects relationships among business and management variables, as well as the applicable contexts of these relationships. The construction of this knowledge graph is aided by our proprietary machine reading system to automate the extraction and synthesis of causal insights from textual knowledge resources such as scientific reports and company reports. Second, the system has a built-in query platform to display the causes and consequences of specified performance outcomes into a machine-readable model. Finally, any subgraphs can be loaded into common statistical software such as Python, R, SAS, Stata, and SPSS to help organizations analyze the data from different silos, and prescribe the optimal intervention based on its predicted outcomes.

The emi® solution will be the first technological solution of digitalized knowledge management to enable enterprise-wide explainable prescriptive analytics concerning enterprise performance management (EPM). 

See a commercial use case [password: KG-BRAIN].