Inventor

JOHNSON MARK EDWARD

AU69 patents

⚠️ This page may combine multiple inventors who share the name “JOHNSON MARK EDWARD”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

JOHNSON MARK EDWARD

ORACLE INT CORP

Compressing neural networks for natural language understanding

Multi-factor modelling for natural language processing

Techniques for out-of-domain (OOD) detection

Transforming natural language to structured query language based on multi- task learning and joint training

Out-of-domain data augmentation for natural language processing

Automatic out of scope transition for chatbot

Context tag integration with named entity recognition models

Reduced training intent recognition techniques

Keyword data augmentation tool for natural language processing

Stop word data augmentation for natural language processing

Noise data augmentation for natural language processing

Entity level data augmentation in chatbots for robust named entity recognition

Using a generative adversarial network to train a semantic parser of a dialog system

Context tag integration with named entity recognition models

Semantic parser including a coarse semantic parser and a fine semantic parser

Automatic out of scope transition for chatbot

Out-of-domain data augmentation for natural language processing

Noise data augmentation for natural language processing

Using a generative adversarial network to train a semantic parser of a dialog system

Using backpropagation to train a dialog system

Techniques for dialog processing using contextual data

Using backpropagation to train a dialog system

Semantic parser including a coarse semantic parser and a fine semantic parser

Implementing a correction model to reduce propagation of automatic speech recognition errors

Enhanced logits for natural language processing

Techniques for out-of-domain (OOD) detection

Distance-based logit value for natural language processing

Distance-based logit value for natural language processing

Techniques for out-of-domain (OOD) detection

Enhanced logits for natural language processing

Calibrating confidence scores of a machine learning model trained as a natural language interface

Techniques for positive entity aware augmentation using two-stage augmentation

Techniques for two-stage entity-aware data augmentation

Data augmentation and batch balancing methods to enhance negation and fairness

Data augmentation and batch balancing methods to enhance negation and fairness

Techniques for augmenting training data for aggregation and sorting database operations in a natural language to database query system

Multi-task model with context masking

Task-oriented dialog suitable for a standalone device

Task-oriented dialog suitable for a standalone device

Reduced training for dialog systems using a database

Streamlining dialog processing using integrated shared resources

Gazetteer integration for neural named entity recognition

Method and system for constraint based hyperparameter tuning

Batching techniques for handling unbalanced training data for a chatbot

Lexical dropout for natural language processing

Fusion of word embeddings and word scores for text classification

Method and system for over-prediction in neural networks

VOICEBOX TECH CORPORATION

Multi-lingual semantic parser based on transferred learning

System and method for generating a multi-lingual and multi-intent capable semantic parser based on automatically generated operators and user-designated utterances relating to the operators

TOUTANOVA KRISTINA NIKOLOVA

Semi-supervised part-of-speech tagging