View source for 18447836. METHOD AND APPARATUS FOR SIMULTANEOUS TRAINING AND CORRECTION OF ARTIFICIAL NEURAL NETWORK AND DATASET simplified abstract (ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE)

=METHOD AND APPARATUS FOR SIMULTANEOUS TRAINING AND CORRECTION OF ARTIFICIAL NEURAL NETWORK AND DATASET=


==Organization Name==

[[:Category:ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE|ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE]]


[[Category:ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE]]

==Inventor(s)==

[[:Category:IL KYU Park of Daejeon (KR)|IL KYU Park of Daejeon (KR)]][[Category:IL KYU Park of Daejeon (KR)]]

[[:Category:Bonki Koo of Daejeon (KR)|Bonki Koo of Daejeon (KR)]][[Category:Bonki Koo of Daejeon (KR)]]

==METHOD AND APPARATUS FOR SIMULTANEOUS TRAINING AND CORRECTION OF ARTIFICIAL NEURAL NETWORK AND DATASET - A simplified explanation of the abstract==


This abstract first appeared for US patent application 18447836 titled 'METHOD AND APPARATUS FOR SIMULTANEOUS TRAINING AND CORRECTION OF ARTIFICIAL NEURAL NETWORK AND DATASET

'''Simplified Explanation''': 

This patent application describes a method for mapping data from different domains to a common joint embedding space using a neural network.

* Training a mapping neural network to create a joint embedding space.
* Generating a prediction matrix for an input dataset using the neural network.
* Creating a merging dictionary to correct the input dataset by merging classes from the prediction matrix.

'''Key Features and Innovation''':

* Utilizes a neural network to map data from diverse domains to a shared embedding space.
* Improves the accuracy and efficiency of data mapping processes.
* Enables the correction of input datasets by merging classes based on prediction matrices.

'''Potential Applications''':

This technology can be applied in fields such as:
* Machine learning
* Data analysis
* Information retrieval

'''Problems Solved''':

* Simplifies the process of mapping data from different domains.
* Enhances the accuracy of data mapping and classification tasks.

'''Benefits''':

* Improved data mapping accuracy.
* Enhanced data classification capabilities.
* Increased efficiency in data analysis tasks.

'''Commercial Applications''''':

Potential commercial applications include:
* Data analytics software
* Machine learning platforms
* Information retrieval systems

'''Questions about Mapping Data to a Joint Embedding Space''':

1. How does the neural network improve the accuracy of data mapping?
2. What are the potential challenges in implementing this technology in real-world applications?


==Original Abstract Submitted==

An embodiment of the present disclosure discloses a method of operating a computing device for mapping data from different domains to a common joint embedding space, and the method of operating a computing device includes training a mapping neural network constituting a joint embedding space using an input dataset, generating a prediction matrix of an input dataset using the mapping neural network, and generating a merging dictionary merging classes from the prediction matrix to correct the input dataset.



[[Category:G06N3/08]]
[[Category:CPC_G06N3/08]]