Classifying relevance of natural language text for topic-based notifications

Aspects of the disclosure include a natural language processing model by which topics of interest within a text are identified, such as by a predictive model that infers (e.g., based on scores associated with a text) a topic of interest associated with the text. The computer system may train or conf...

Full description

Saved in:
Bibliographic Details
Main Authors Leon Ortiz, Ana Maria, Tiollier, Jérôme, Tissier-Seta, Jean Philippe, Caschera, Elisabetta, Gerbaud, Thomas, Voisin, Céline, Marino, Jérôme, Nasr, Sabine
Format Patent
LanguageEnglish
Published 11.06.2024
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:Aspects of the disclosure include a natural language processing model by which topics of interest within a text are identified, such as by a predictive model that infers (e.g., based on scores associated with a text) a topic of interest associated with the text. The computer system may train or configure the prediction model, such as a machine learning model, to facilitate identification of topics of interest based on inputs, like one or more chunks of text, such as by keywords or phrases or combinations of keywords and associated metrics for nearness or frequency. The computer system may determine a measure of predicted impactfulness of the content item in relation to a topic of interest identified for the content item and determine whether to generate a notification transmitted to client devices of users having indicated the topic as of interest.
Bibliography:Application Number: US202217744327