17960022. UNKNOWN OBJECT SUB-CLASS IDENTIFICATION simplified abstract (International Business Machines Corporation)

From WikiPatents
Jump to navigation Jump to search

UNKNOWN OBJECT SUB-CLASS IDENTIFICATION

Organization Name

International Business Machines Corporation

Inventor(s)

Vijay Sundaresan of Markham (CA)

UNKNOWN OBJECT SUB-CLASS IDENTIFICATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 17960022 titled 'UNKNOWN OBJECT SUB-CLASS IDENTIFICATION

Simplified Explanation

The present specification describes a computer-implemented method for determining the class of an unknown object.

  • First comparison test to determine if unknown object is of a first sub-class.
  • If not, check if unknown object is an instance of a second sub-class.
  • If there are additional sub-classes, refrain from assuming the unknown object is of a particular sub-class.

---

      1. Potential Applications

This technology could be applied in various fields such as image recognition, pattern recognition, and data classification.

      1. Problems Solved

This technology helps in accurately categorizing unknown objects into specific sub-classes, improving classification accuracy.

      1. Benefits

- Enhanced object classification accuracy - Efficient handling of unknown objects - Improved decision-making based on object classification

      1. Potential Commercial Applications
        1. Optimizing Object Classification for Improved Decision Making
      1. Possible Prior Art

No known prior art at this time.

---

      1. Unanswered Questions
        1. How does this technology handle complex object structures with multiple sub-classes?

This article does not delve into the specifics of handling complex object structures with multiple sub-classes. Further research may be needed to understand the scalability of this technology in such scenarios.

        1. Are there any limitations to the size or type of objects that can be classified using this method?

The article does not address any limitations regarding the size or type of objects that can be classified using this method. It would be beneficial to explore the boundaries of this technology in terms of object diversity and complexity.


Original Abstract Submitted

The present specification describes a computer-implemented method. A first comparison test is executed to determine whether an unknown object is of a first sub-class of a class of objects. Responsive to determining that the unknown object is not of the first sub-class, it is determined whether the unknown object is an instance of a second sub-class by determining whether there are additional sub-classes other than the first sub-class and a second sub-class. Responsive to determining that there are additional sub-classes, the second code fragment executes while refraining from assuming the unknown object is of a particular sub-class.