Multilingual text quality scoring

Prerequisites

• You can use the workflow feature only after you purchase the unstructured data feature for the current instance.

• A workflow has been created. For more information, see Create a workflow.

Background information

• The multilingual text quality scoring operator processes files by parsing only the field values in the metadata table.

• By default, the multilingual text quality scoring operator uses 1 Core and 2 GB of resources with the system built-in model. With a multi-modal Large Language Model (LLM), it uses 0.5 Core and 1 GB of resources.

Permissions

Project administrators, developers, and analysts of a Basic project, and users in custom project roles with the Dataset-Use permission, can use any dataset in the project as an input or output dataset for the operator.

Procedure

1. On the Dataphin home page, go to the top menu bar and choose Develop > Data Development.

2. In the top menu bar of the Develop page, select Project.

3. In the navigation pane on the left, choose Data Processing > Workflow. In the Workflow list on the left, click your target workflow to open the workflow canvas. From the Operator Library on the left, drag the Multilingual Text Quality Scoring operator to the canvas. Then, on the Multilingual Text Quality Scoring panel on the right, configure the following parameters.

   Parameter		Description
   Step Name		Enter a name for the operator. The name can contain any characters and be up to 256 characters long.
   Input Configuration	Dataset	Select the dataset to process, which can be a mixed dataset or a table dataset. After you select a dataset, select a version. By default, the first version of the dataset in ascending order is used. If this operator is a downstream operator, the selected dataset is usually the same as the output dataset of the upstream operator.
   	Input Field	Select a character-type field from the input dataset that stores the text content.
   	Filter Condition (Optional)	Use SQL syntax. Enter only the filter condition that follows the WHERE clause. System global variables, such as the data timestamp ${bizdate}, are supported.
   Model Configuration	Processing Model	You can select Default (system Built-in Model) (multilingual-e5-small-aligned-quality) or Multimodal Large Language Model. If you select Multimodal Large Language Model, you must specify the Model for the current operator. You can only select models from Intelligent Application Management > Model Configuration that have a model type that includes Text and are Enabled.
   Output Configuration	Dataset	You can select Source Dataset or Other Dataset. If the metadata table of the source dataset does not have a primary key, the output dataset defaults to Other dataset. If you select Other Datasets, specify a mixed dataset or a table dataset and its version as the output dataset. By default, the first version of the dataset in ascending order is used. You can also select other versions of the source dataset. When you select Source Dataset, the metadata generated by the operator is written to the metadata table that corresponds to the same version of the source dataset.
   	Loading Policy	If the metadata table of the selected output dataset has a primary key, Update on Primary Key Conflict is selected by default. If no primary key exists, Append Data is selected by default. Append Data: Appends new data directly to the destination table. A constraint violation error is reported if a primary key conflict occurs. Update on Primary Key Conflict: If a primary key conflict or constraint violation occurs, the old data in the entire row with the duplicate primary key is deleted before the new data is inserted. Overwrite: Deletes the data from the destination table before writing the new data. Note If the output dataset is set to Source Dataset, the only supported loading policy is Update on Primary Key Conflict.
   	Field Mapping	The multilingual text quality scoring operator has the following fixed output field: quality_score: The text quality score. The score ranges from 0 to 1. A higher score indicates better quality. A score greater than 0.5 is generally considered good. By default, the operator's output fields are mapped to fields with the same names in the output dataset. You can also manually select destination fields for custom mapping. Pass-through fields from the source dataset can also be mapped to the destination dataset. Manage field mappings in the following ways: Add Output Field: Click to add a new empty row for field mapping. This feature is available only when you select Other Datasets for the output dataset. Batch Mapping: Click to choose batch manual mapping, same-name mapping, or same-row mapping. This feature is available only when you select Other Datasets for the output dataset. Delete Mapping: Click the Delete icon to delete the field mapping for the corresponding row. The mapping for the operator's fixed output field cannot be deleted.

4. After you complete the configuration, click Save in the top menu bar of the canvas.

   After you configure an operator, you can add upstream or downstream operators to build a complete workflow.