Patent · US Active

System and method for generating contextualized text using a character-based convolutional neural network architecture

US12086542B2 · kind B2 · utility

0Cited by
13References
16Claims
0Family size

Assignee

Inventors

Key dates

Filing dateApr 6, 2022
Grant dateSep 10, 2024
Priority date
Expiry dateMar 14, 2043

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N3/09
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A system for generating a contextualized text using a character-based convolutional neural network architecture (CNN) is disclosed. The system includes a memory, a character-based CNN engine having executable modules and a processor configured to execute the character-based CNN engine and the executable modules for generating contextualized text using a character-based convolutional neural network architecture (CNN). The executable modules includes a data processing module for pre-processing an input conversational text data, a corpus builder unit for developing a corpus of most frequent n-grams, a dataset generation module for performing logarithmic mapping of a frequency of a respective n-gram in the corpus for a dataset prepared for training, a training module for training an artificial intelligence (AI) based model with datasets for a predetermined range of epochs and a prediction module for executing the trained AI-based model for predicting characters contextually till an end-of-sentence (EOS) token is found.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.