18521805. CONTEXT TAG INTEGRATION WITH NAMED ENTITY RECOGNITION MODELS simplified abstract (Oracle International Corporation)

CONTEXT TAG INTEGRATION WITH NAMED ENTITY RECOGNITION MODELS

Organization Name

Oracle International Corporation

Inventor(s)

Duy Vu of Melbourne (AU)

Tuyen Quang Pham of Melbourne (AU)

Cong Duy Vu Hoang of Wantirna South (AU)

Srinivasa Phani Kumar Gadde of Fremont CA (US)

Thanh Long Duong of Seabrook (AU)

Mark Edward Johnson of Castle Cove (AU)

Vishal Vishnoi of Redwood City CA (US)

CONTEXT TAG INTEGRATION WITH NAMED ENTITY RECOGNITION MODELS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18521805 titled 'CONTEXT TAG INTEGRATION WITH NAMED ENTITY RECOGNITION MODELS

Simplified Explanation

The abstract describes a method for using context tags in named-entity recognition (NER) models. Here is a simplified explanation of the patent application:

• Receiving an utterance
• Generating embeddings for words of the utterance
• Generating a regular expression and gazetteer feature vector for the utterance
• Generating a context tag distribution feature vector for the utterance
• Concatenating or interpolating the embeddings with the regular expression and gazetteer feature vector and the context tag distribution feature vector to generate a set of feature vectors
• Generating an encoded form of the utterance based on the set of feature vectors
• Generating log-probabilities based on the encoded form of the utterance
• Identifying one or more constraints for the utterance

Potential Applications: - Improving accuracy and efficiency of named-entity recognition systems - Enhancing natural language processing tasks

Problems Solved: - Enhancing the performance of NER models by incorporating context tags - Improving the identification of named entities in text data

Benefits: - Increased accuracy in identifying named entities - Better understanding of the context in which entities appear in text

Potential Commercial Applications: - Text analysis tools for businesses - Information extraction systems for research purposes

Possible Prior Art: - Existing NER models that do not utilize context tags

Unanswered Questions: 1. How does the method handle ambiguous entities in the text data? 2. What is the computational complexity of the proposed approach compared to traditional NER models?

Original Abstract Submitted

Techniques are provided for using context tags in named-entity recognition (NER) models. In one particular aspect, a method is provided that includes receiving an utterance, generating embeddings for words of the utterance, generating a regular expression and gazetteer feature vector for the utterance, generating a context tag distribution feature vector for the utterance, concatenating or interpolating the embeddings with the regular expression and gazetteer feature vector and the context tag distribution feature vector to generate a set of feature vectors, generating an encoded form of the utterance based on the set of feature vectors, generating log-probabilities based on the encoded form of the utterance, and identifying one or more constraints for the utterance.