Dynamic attribute split for NLP-oriented AI data analytics
OwnerMaxim Chuprunovinventor
Life of this patent
Estimating the term…
Abstract
Systems, methods, and computer algorithms disclosed herein enable dynamic adjustment of data structures in data sources used for Artificial Intelligence (AI) assisted data analytics, especially when Natural Language Processing (NLP) is used. This method includes a dynamic creation of attribute fields, definition of their names, and content based on existing datasets. Such transformation of the underlying data aims to enhance performance of NLP-based data processing by making it better match human language.
Claims3
A method for attribute splitting to enhance the data analysis capabilities of AI models, comprising techniques for identifying and transforming distinct attribute data within datasets.
A system for attribute splitting integrated with AI models that facilitates improved efficiency and accuracy in data analysis tasks, particularly in applications involving Natural Language Processing (NLP).
A novel algorithm for increasing the success rate of deriving correct SQL logic from natural language input, utilizing an attribute splitting technique in AI models.
Dependencies are inferred from the claim text.