Abstract: Embodiments of the present invention provide methods, systems, and/or the like for versioning a graph representation in a graph data structure. In accordance with one embodiment, a method is provided comprising: conducting a plurality of iterations involving: validating a first data source comprising a new version of data based on a schema from a plurality of schemas in which each schema corresponds to a graph representation found in a graph data structure; and identifying errors in the first source based on the validating of the source; identifying an applicable schema as a schema producing fewer errors than at least one other schema; comparing the first source with a second source comprising a previous version of the data to identify a difference; generating a query for the difference based on the applicable schema; and providing the query for execution to migrate the difference into the graph representation.

Abstract: Aspects of the present invention provide methods, apparatuses, systems, computing devices, computing entities, and/or the like for detecting prejudice bias in machine-learning models and/or data sets used in training, testing, and/or validating the models. In accordance various aspects, a method is provided comprising: receiving a data set used for training, testing, and/or validating a model that comprises data instances; generating, using a classification model, a prediction of applicability for each sub-category of a plurality of sub-categories for each bias category of a plurality of bias categories for each data instance; determining that a particular sub-category for a particular bias category is applicable to a proportion of the data set, wherein predictions of applicability for the particular sub-category generated for the proportion of the data set satisfies a threshold; and determining, based on the proportion, that the data set has a prejudice bias with respect to the particular bias category.