Abstract
This publication discloses methods and systems for classifying prior sealed decision records into a plurality of reuse classifications based on structural relationships between contextual positions, where each reuse classification imposes different mode- specific requirements and produces categorically different downstream effects that are sealed at classification time. In one embodiment, a post-retrieval classification process evaluates a candidate sealed decision record relative to a current decision context and selects one reuse classification from among a plurality of reuse classifications. Examples of possible reuse classifications include, but are not limited to, substitutive reuse where the prior record directly satisfies the current request, advisory reuse where the prior record informs but does not satisfy the current request, and evidentiary reuse where evidence from the prior record transfers to the current context without record-level satisfaction. In one embodiment, a key structural property is classification immutability: once a reuse classification is determined, the downstream effects associated with that classification cannot be altered by any subsequent processing step, ensuring sealed reuse classification throughout execution. Different reuse classifications apply different subsets of compatibility dimensions — evaluated as a conjunctive compatibility evaluation including positional compatibility, evaluation-context compatibility, and specification-version compatibility — such that a stricter classification requires more dimensions to be satisfied than a less strict classification. In one embodiment, the reuse classification taxonomy is open-ended: the system accepts a plurality of reuse classifications with at least two but does not constrain the taxonomy to a prescribed set, enabling mode- specific reuse requirements and tiered decision record reuse across organizational contexts. The approach relates to regulatory compliance automation systems in which prior approval decisions must be reused across organizational contexts with differentiated applicability — for example, procurement approval reuse where a prior approval may directly satisfy, merely inform, or provide evidence for a new procurement request depending on the structural relationship between the approval context and the request context. The disclosed approaches establish reuse mode classification as a structural determination based on the nature of the relationship between contextual positions, distinguishing the approach from similarity-based retrieval, freshness-based categorisation, and administrative designation of prior decision application modes.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Winchester, Jayson, "Case-Based Reasoning Post-Retrieval Classification — Immutable Reuse Mode Assignment", Technical Disclosure Commons, (February 19, 2026)
https://www.tdcommons.org/dpubs_series/9351