18522028. MODEL TRAINING METHOD AND APPARATUS, SERVICE PROCESSING METHOD AND APPARATUS, STORAGE MEDIUM, AND DEVICE simplified abstract (Alipay (Hangzhou) Information Technology Co., Ltd.)

MODEL TRAINING METHOD AND APPARATUS, SERVICE PROCESSING METHOD AND APPARATUS, STORAGE MEDIUM, AND DEVICE

Organization Name

Alipay (Hangzhou) Information Technology Co., Ltd.

Inventor(s)

Weiqiang Wang of Hangzhou (CN)

Jinzhen Lin of Hangzhou (CN)

Zhenzhe Ying of Hangzhou (CN)

Lanqing Xue of Hangzhou (CN)

MODEL TRAINING METHOD AND APPARATUS, SERVICE PROCESSING METHOD AND APPARATUS, STORAGE MEDIUM, AND DEVICE - A simplified explanation of the abstract

This abstract first appeared for US patent application 18522028 titled 'MODEL TRAINING METHOD AND APPARATUS, SERVICE PROCESSING METHOD AND APPARATUS, STORAGE MEDIUM, AND DEVICE

Simplified Explanation

The present specification discloses a model training method and apparatus, a service processing method and apparatus, a storage medium, and a device. The model training method includes: obtaining a historical conversation; determining a target conversation content from the historical conversation; inputting the historical conversation into a to-be-trained feature extraction model for the feature extraction model to determine a conversation content feature corresponding to the target conversation content as a first feature based on a conversation content other than the target conversation content in the historical conversation, and to determine a conversation content feature corresponding to the target conversation content as a second feature based on the target conversation content; and training the feature extraction model with an optimization goal of reducing a deviation between the first feature and the second feature, where the trained feature extraction model is used to determine an output conversation content feature corresponding to each input conversation content, and send the output conversation content feature for a receiving end to perform service processing based on the received output conversation content feature.

• Model training method and apparatus:

   - Obtaining historical conversation data
   - Determining target conversation content
   - Training feature extraction model to reduce deviation between features

• Service processing method and apparatus:

   - Using trained model to determine output conversation content feature
   - Sending output feature for service processing at receiving end

• Storage medium and device:

   - Storing historical conversation data
   - Implementing trained model for service processing

Potential Applications

This technology can be applied in customer service chatbots, virtual assistants, and automated messaging systems.

Problems Solved

This technology helps improve the accuracy and efficiency of automated conversation systems by training models to better understand and generate conversation content.

Benefits

- Enhanced user experience in interacting with automated systems - Increased productivity and cost savings for businesses using automated conversation systems

Potential Commercial Applications

Optimizing customer service operations, streamlining communication processes, and improving user engagement in various industries.

Possible Prior Art

One possible prior art could be the use of machine learning models for natural language processing in chatbots and virtual assistants.

How does this technology handle multi-language conversations?

This article does not address how the model training method and apparatus handle conversations in multiple languages.

What is the computational cost associated with training the feature extraction model?

The article does not provide information on the computational resources required for training the feature extraction model.

Original Abstract Submitted

The present specification discloses a model training method and apparatus, a service processing method and apparatus, a storage medium, and a device. The model training method includes: obtaining a historical conversation; determining a target conversation content from the historical conversation; inputting the historical conversation into a to-be-trained feature extraction model for the feature extraction model to determine a conversation content feature corresponding to the target conversation content as a first feature based on a conversation content other than the target conversation content in the historical conversation, and to determine a conversation content feature corresponding to the target conversation content as a second feature based on the target conversation content; and training the feature extraction model with an optimization goal of reducing a deviation between the first feature and the second feature, where the trained feature extraction model is used to determine an output conversation content feature corresponding to each input conversation content, and send the output conversation content feature for a receiving end to perform service processing based on the received output conversation content feature.