Systeme und Verfahren zur Einbettung und Gruppierung von Dokumenten auf der Basis von Maschinellem Lernen mit Nachfolgender Aktionsreihung und -Benachrichtigung

Details

Veröffentlichungs-Nr.

EP4752754

Anmeldetag

26. November 2025

Veröffentlichung

3. Juni 2026

Rechtsraum

EP

IPC

G06F16/906

Abstract

This application describes, amongst other things, example methods and systems using machine-learning components having knowledge of a plurality of numerical embeddings corresponding to a plurality of documents, where the plurality of numerical embeddings relates to a set of similarity clusters. An example method includes obtaining a constraint indicating a period of time and a type of subject matter and, using the constraint, obtaining a set of documents and corresponding metadata. A set of numerical embeddings are generated for the set of documents. A set of updated similarity clusters are generated by updating the set of similarity clusters to incorporate the set of numerical embeddings. A set of potential actions are identified based on links between entity information and the set of updated similarity clusters. The set of potential actions are ranked, and a notification is provided indicating top-ranked actions.

Anmelder

Firma

Tempus AI, Inc.

Land

🇺🇸 USA