Custom Machine Translation Manager (Custom Glossary Model List)

What is custom glossary setting/custom glossary model?

The process of wrapping the standard model of MT learning service or the custom trained model registered in this system using a glossary to automatically output the translation defined in the glossary is called “Custom glossary setting” in this system (sometimes called "Glossary wrapping"). As the result of custom glossary setting, a custom glossary model is created. Once the custom glossary setting has been completed, the models set with glossaries can be used in Quick MT/PE of this system.

Information in Custom Glossary Model List screen

On this screen, a list of custom glossary models registered in the organization is displayed. You can also search for a custom glossary model by specifying conditions. To add a new custom glossary model, click on "+ Register" button at the top of the screen.

• Search conditions area
  
  • Machine Translation Learning Service: Specifies the service used for custom glossary setting of the model. The following three types of MT learning services can be used in this system.
    
    • Google AutoML
    • Kawamura NMT
    • Amazon Translate
  • Status: See the section below for changes in the status of the custom glossary models. In Quick MT/PE and LAC, only models with a status of “Custom Glossary Setting Completed” are available.
  • Model Status: Only models with "Enabled" status are available in Quick MT/PE.
• List of custom glossary models
  
  • Custom Glossary Model Name: Click to go to Custom glossary model details screen.
  • MT Learning Service
  • Custom Trained Model Name: If you have set a glossary against a custom trained model, the name of it is displayed. If a standard model was used, this field is blank.
  • Enabled: "○" is displayed when the model is enabled.
  • Start Date/Time, End Date/Time: Start and end date/time of the custom glossary setting.
  • Status
  • Evaluation Value: If you have already performed a model evaluation for the custom glossary model, the BLEU value is displayed out of the four quality indicators. This field is blank if the model evaluation has not been performed. There will be two evaluation values displayed when a standard model is used for the glossary setting, and will be three values when a custom trained model is used.
  • Comment: Displays the comments entered in Custom glossary model details screen.
•  

Status of a custom glossary model

The status of a custom glossary model can be one of the following 10 types.

Deletion of custom glossary models

You can select one or more models and delete them at once. Select the check box of the target model from the list, and click [Delete] icon in Batch Processing menu. Please note that deleting a model has the following effects.

• LAC (Custom Machine Translation Management Menu): The custom glossary model is deleted and no longer appears in the list.
• Quick MT: When data for a custom glossary model is deleted, it no longer appears in the pull-down list of the available custom glossary models for each MT engine.
• Quick PE: When data for a custom glossary model is deleted, it no longer appears in the pull-down list of the available custom trained models for each MT engine. Also, if there is a project that selects the deleted custom glossary model, it is possible to reopen File Editor screen of the files in that project, but re-translation will result in an error.

Model evaluation of custom glossary models

Click the name of each custom glossary model in the list to go to the custom glossary model details screen where you can evaluate the model. [Evaluation Value] column in the list shows only the BLEU value out of the four metrics. You can check all evaluation values on the custom glossary model details screen.

Before the “→” sign is the evaluation value of the standard model (*) of MT learning service used for custom glossary, and the latter is the evaluation value of the custom glossary model. If the values of BLUE, NIST, and RIBES changes larger, and WER changes smaller, it can be interpreted that effective customization has been implemented by custom glossary setting. You can also check the detailed meaning of each metric in this blog post (Japanese).

• BLEU
• NIST
• RIBES
• WER