International business machines corporation (20240104369). KNOWLEDGE EXPANSION FOR IMPROVING MACHINE LEARNING simplified abstract

KNOWLEDGE EXPANSION FOR IMPROVING MACHINE LEARNING

Organization Name

international business machines corporation

Inventor(s)

Dinesh C. Verma of New Castle NY (US)

Franck Vinh Le of West Palm Beach FL (US)

Michele Merler of New York City NY (US)

Dhiraj Joshi of Edison NJ (US)

SUPRIYO Chakraborty of White Plains NY (US)

Seraphin Bernard Calo of Cortlandt Manor NY (US)

KNOWLEDGE EXPANSION FOR IMPROVING MACHINE LEARNING - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240104369 titled 'KNOWLEDGE EXPANSION FOR IMPROVING MACHINE LEARNING

Simplified Explanation

The patent application describes a system that can train a neural network on a base set of knowledge, deploy the neural network on a new data set, generate new instances of knowledge using the deployment, and validate the new instances of knowledge.

• The system receives an existing base set of knowledge.
• It trains a neural network on the base set of knowledge.
• The trained neural network is deployed on a new data set.
• Using the deployment, the system generates instances of new knowledge.
• The system then validates the instances of new knowledge.

Potential Applications

The technology described in this patent application could be applied in various fields such as:

• Data analysis
• Predictive modeling
• Natural language processing

Problems Solved

This technology helps in:

• Automating knowledge generation
• Improving decision-making processes
• Enhancing data processing efficiency

Benefits

The benefits of this technology include:

• Faster generation of new knowledge
• Increased accuracy in predictions
• Scalability for handling large datasets

Potential Commercial Applications

This technology could be utilized in industries such as:

• Healthcare for medical diagnosis
• Finance for risk assessment
• E-commerce for personalized recommendations

Possible Prior Art

One possible prior art for this technology could be the use of neural networks in data analysis and predictive modeling.

Unanswered Questions

How does the system handle noisy data during the training process?

The system's ability to filter out noise and irrelevant information could impact the accuracy of the generated knowledge.

What is the computational cost of deploying the neural network on a new data set?

Understanding the computational resources required for deployment is crucial for assessing the scalability of the system.

Original Abstract Submitted

a system may receive an existing base set of knowledge, train a neural network on the base set of knowledge, deploy the neural network on a new data set, generate, using the deployment, instances of new knowledge, and validate the instances of new knowledge.