18154149. DEEP SYMBOLIC VALIDATION OF INFORMATION EXTRACTION SYSTEMS simplified abstract (INTERNATIONAL BUSINESS MACHINES CORPORATION)

From WikiPatents
Jump to navigation Jump to search

DEEP SYMBOLIC VALIDATION OF INFORMATION EXTRACTION SYSTEMS

Organization Name

INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor(s)

Alfio Massimiliano Gliozzo of Brooklyn NY (US)

Sarthak Dash of Jersey City NJ (US)

Michael Robert Glass of Bayonne NJ (US)

Mustafa Canim of Ossining NY (US)

DEEP SYMBOLIC VALIDATION OF INFORMATION EXTRACTION SYSTEMS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18154149 titled 'DEEP SYMBOLIC VALIDATION OF INFORMATION EXTRACTION SYSTEMS

Simplified Explanation

The patent application describes a system that includes a memory and a processor, which work together to execute computer-executable components. The system has three main components: a receiving component, a relation extraction component, and a training component.

  • The receiving component is responsible for receiving a corpus of data.
  • The relation extraction component takes the corpus of data and generates noisy knowledge graphs from it.
  • The training component acquires global representations of entities and relations by training from the output of the relation extraction component.

Potential applications of this technology:

  • Natural language processing: The system can be used to extract relations and knowledge from large amounts of text data, which can be valuable in various applications such as information retrieval, question answering, and text summarization.
  • Knowledge graph construction: The generated noisy knowledge graphs can be used to build and update knowledge graphs, which can be used in various domains like semantic search, recommendation systems, and knowledge-based reasoning.

Problems solved by this technology:

  • Efficient extraction of relations: The system automates the process of extracting relations from a corpus of data, saving time and effort compared to manual extraction.
  • Handling noisy data: The system is capable of generating noisy knowledge graphs, which can handle the inherent noise and ambiguity present in natural language data.

Benefits of this technology:

  • Scalability: The system can handle large amounts of data, making it suitable for processing big data.
  • Automation: The system automates the extraction of relations, reducing the need for manual effort.
  • Knowledge representation: The acquired global representations of entities and relations can provide a structured representation of knowledge, enabling better understanding and utilization of the data.


Original Abstract Submitted

A system comprises a memory that stores computer-executable components; and a processor, operably coupled to the memory, that executes the computer-executable components. The system includes a receiving component that receives a corpus of data; a relation extraction component that generates noisy knowledge graphs from the corpus; and a training component that acquires global representations of entities and relation by training from output of the relation extraction component.

Retrieved from "http://wikipatents.org/index.php?title=18154149._DEEP_SYMBOLIC_VALIDATION_OF_INFORMATION_EXTRACTION_SYSTEMS_simplified_abstract_(INTERNATIONAL_BUSINESS_MACHINES_CORPORATION)&oldid=25742"