Device and method for machine learning

[0006]In accordance with an example embodiment of the present invention, a computer-implemented method for grouping target entities into clusters provides that a base association in which a cluster is associated with each of the target entities is determined in a computation step for the target entities as a function of an association for entities, inference rules being determined as a function of the association for entities and as a function of the base association, each of the inference rules defining an association of entities with one of the clusters, an altered association being determined as a function of the association for entities and the inference rules, a check being made as to whether a difference between the base association and the altered association falls below a threshold value, and when the difference falls below the threshold value an association of the target entities with the clusters being output or stored, and otherwise a feedback value being determined as a function of the difference, and an association being determined, as a function of the association for entities and the feedback value, which replaces the association for a new execution of the computation step. The associations may be represented by knowledge graphs. In this way, for a given knowledge graph and a set of predefined target entities, a set of entities of the knowledge graph are better grouped into clusters. This represents an iterative process in which the target entities are grouped i

It is difficult to find valid descriptions for predefined diffuse clusters.

Since the clusters are determined based on statistical tes