18283411. HYBRID MODEL CREATION METHOD, HYBRID MODEL CREATION DEVICE, AND RECORDING MEDIUM simplified abstract (Panasonic Intellectual Property Management Co., Ltd.)

HYBRID MODEL CREATION METHOD, HYBRID MODEL CREATION DEVICE, AND RECORDING MEDIUM

Organization Name

Panasonic Intellectual Property Management Co., Ltd.

Inventor(s)

Yao Zhou of Singapore (SG)

Athul M. Mathew of Singapore (SG)

Ariel Beck of Singapore (SG)

Chandra Suwandi Wijaya of Singapore (SG)

Nway Nway Aung of Singapore (SG)

Khai Jun Kek of Singapore (SG)

Yuya Sugasawa of Osaka (JP)

Jeffry Fernando of Osaka (JP)

Yoshinori Satou of Osaka (JP)

Hisaji Murata of Osaka (JP)

HYBRID MODEL CREATION METHOD, HYBRID MODEL CREATION DEVICE, AND RECORDING MEDIUM - A simplified explanation of the abstract

This abstract first appeared for US patent application 18283411 titled 'HYBRID MODEL CREATION METHOD, HYBRID MODEL CREATION DEVICE, AND RECORDING MEDIUM

Simplified Explanation

The abstract describes a patent application for a method that involves pooling multiple models that predict categories of input data, including at least one model trained by machine learning. Hybrid model candidates are then created by selecting and combining two or more models from the pooled models, and one hybrid model candidate is selected based on comparison.

• The method involves pooling multiple models that predict categories of input data.
• At least one of the models is trained using machine learning.
• Hybrid model candidates are created by combining two or more models from the pooled models.
• One hybrid model candidate is selected based on comparison with others.

Potential Applications

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

• Predictive analytics
• Image recognition
• Natural language processing

Problems Solved

This technology helps in:

• Improving accuracy of category predictions
• Enhancing the performance of machine learning models
• Streamlining the process of model selection

Benefits

The benefits of this technology include:

• Increased efficiency in predicting categories of input data
• Enhanced accuracy in classification tasks
• Optimized model selection process

Potential Commercial Applications

The technology could be utilized in:

• E-commerce for product categorization
• Healthcare for disease diagnosis
• Finance for fraud detection

Possible Prior Art

One possible prior art for this technology could be the use of ensemble learning techniques in machine learning models to improve prediction accuracy.

How does this technology compare to existing methods of model selection in machine learning?

This article does not provide a direct comparison to existing methods of model selection in machine learning.

What are the potential limitations of this technology in real-world applications?

The article does not address the potential limitations of this technology in real-world applications.

Original Abstract Submitted

First, a plurality of models that predict categories of input data are pooled. At least one of the plurality of models is a model trained by machine learning. Next, each of a plurality of hybrid model candidates that judge the categories are created by selecting and combining two or more models from among the plurality of pooled models. Then, by comparing the plurality of hybrid model candidates, one of the plurality of hybrid model candidates is selected as a hybrid model.