What is Enterprise Category Theory
A maturing technology is emerging from powerful mathematics: Category Theory. Category Theory has evolved from concepts to study algebraic topology to a means of formal representation of abstract processes.
During research funded by the US Federal Defense Advanced Research Projects Agency (DARPA), and NIST (NIST SBIR) the principals and founders at Conexus have advanced Category Theory knowledge leading to embedding functions in software. The advanced knowledge has advanced academically as Applied Category Theory (ACT). ACT relates the mathematical principles and robust representational formalisms for working in other domains like the health sciences, information processing, and research & development (R&D). The emerging solutions combine support and sophisticated features, services and software for ACT. These offerings enable enterprises to resolve hard problems with insights from ACT.
ACT deals with the extension and grounding of Category Theory into domain knowledge in applied sciences and engineering. Enterprise Category Theory (EntCT) brings practical extensions to ACT for applied R&D, and information processing for organizations in industry, health care, government, and the sciences. EntCT deals with the Domains of Dirty Data, dynamic datafeeds and sources, and practical integration of ACT into information technology operations. Implementation and deployment of EntCT resolves issues in the quality attributes of data for data sciences, machine learning, artificial intelligence developments and deployment, and complex databases.
EntCT drives analysis and systems design insights to accelerate problem resolution. EntCT techniques gain users deeper insights into data. EntCT relieves constraints and accelerates analysis for multiple data models (such as graph databases), multiple datastores, and improving data models. EntCT deals with metadata, complex data lineage, and legacy datastores. Knowledge embedded in software improves data migration, reduction, and builds confidence in datastores for enterprise efficiencies and effective analytics. Business benefits from these gains can be recognized in early and sustaining phases of large initiatives.
The general goals of Enterprise Category Theory applied to data sciences and machine learning are to improving the quality (measured by evaluating accuracy, validity, verifiable, and correlations of results) and useful accuracy of the results for data sciences/analytics and machine learning.
The uses of Enterprise Category Theory apply to:
- Definition – using formal representation to reduce ambiguity, maintain data integrity across transformation, and establish uniformity across multiple representational forms in larger efforts
- Statement – using formal representations to clarify lineage, processing, and constraints for processing events, flows, and data transformations and uses
- Analysis – using metrics and logical reasoning enabled by having a consistent formal representation of definitions and statements for data and operations on data
- Refinement of data and processes – using information processing and analytical processes to refine the quality characteristics of data, to attempt to ensure valid processing of data, and to enable dynamic updating with verifiable integrity checking steps for changes induced or deduced from subsequent data or processing events
Born of breakthroughs in Mathematics at MIT. Productized with PhDs from Harvard. Built for scale with PhDs from Carnegie Mellon. Conexus industry deployments have demonstrated value unavailable with other approaches. Conexus clients avoid the risks of catastrophic failure, limitless costs, and interminable timelines that are featured as part of most all data integration and migration projects. We have productized a platform-as-a-service that enables open-market opportunities for developers.