Configuration procedure
1. In the OpenSearch Industry Algorithm Edition console, go to the Instance Management page and click Create Instance.
2. On the purchase page, set Edition to Industry Algorithm Edition and Industry Type to Education. For Cluster Preference, select exclusive cluster - General-purpose/Compute-optimized/Storage-optimized. The other settings are the same as for the General-purpose type.
3. Confirm the order:
On the Confirm Order page, verify the configuration details, such as region, application name, edition type, industry type, purchase specifications, storage capacity, and computing resources. After you confirm the billing information, select the service agreement and click Buy Now.
4. After the application is created, find it on the Application Management page and click Configure.
Then, configure the application by completing the following steps in order: Feature Selection --> Application Schema --> Index Schema --> Data Source --> Complete.
(1) In the Feature Selection step, select "Educational Q&A Search" for the vertical category. You can select the required template features. By default, all features are selected. Then, click Next.
The template features include query analysis with a dedicated education thesaurus, a dedicated education sort policy, and drop-down suggestions.
(2) In the Application Schema step, this guide uses the Manually Create method as an example. After you complete the configuration, click Next.
The main table is automatically generated when you create an application and cannot be manually created. The console allows you to add only secondary tables.
Set the table name to main and select it as the main table. Add the following fields: id (INT type, set as the primary key), title (TEXT type, select Question Title for the field tag), content (TEXT type, select Question Content for the field tag), and desc (LITERAL type). Assign the corresponding field tags to each field. Note: Field tags help you quickly generate the industry-enhanced index schema, query analysis rules, and sort policies. You must add a field tag to the corresponding field in the application schema.
(3) Index Schema:
This page has two tabs: Default Text Index and vector index. You need to set the index tag, included fields, and analysis method for each index. You also need to set the corresponding fields in the Attribute Fields list and Default Display Fields list. When finished, click Next.
Note: Index tags are used to quickly generate query analysis rules and sort policies. You must add an index tag to the corresponding index in the index schema. The mapping is as follows: 1. The fields configured with the Question Title and Question Content field tags in the application schema must be set as a composite index. Add the "default text index" index tag to this index, and set the analysis method to "Industry - Education Question Search". 2. For the field configured with the Question Title field tag, you must add the "vector index" index tag and set the analysis method to "Vector - K-12 Education Vector".
(4) Data Source: OpenSearch supports cloud data sources such as ApsaraDB RDS for MySQL, MaxCompute, and PolarDB.
Note: You can also upload data to OpenSearch using an API.
(5) Finalize configuration:
After the configuration is complete, a message appears indicating that the configuration for the Industry-enhanced Edition application is complete. You can now perform the following operations: 1. Configure query analysis using the query analysis feature pack of the selected industry template. 2. Configure a sort policy using the selected sort policy model. 3. Configure and use the drop-down suggestions feature. Click Return to Application List to go back to the Application Management page.
Query test
Search test
After the index is built, you can test the index's industry-specific text analysis on the Feature Extensions > Search Test page. Enter a query clause in the query input box, configure the clause and parameters, and then click Search to view the results.
For the default index, enter the query clause: default:'Education'Vector search
To test the vector search feature, follow these steps:
1. On the Search Algorithm Center > Retrieval Configuration > Query Analysis Configuration page, configure the text vectorization feature for the default sys_default query analysis rule from the industry template.
Click the sys_default query analysis rule to edit it. In the Add Rule dialog box, confirm that the industry type is Education, select the Text Vectorization checkbox in the feature selection area, and click OK.
2. On the Feature Extensions > Search Test page, add the required parameters to run a test.
Required parameters
-
query=default:'Education'
-
qp=sys_default
-
raw_query=Education
Note:
-
The vector index takes effect only after you configure the text vectorization feature for query analysis.
-
When you configure query analysis, ensure that the following conditions are met: The index must use the "Industry - Education Question Search" analyzer. The industry must be set to "Education". For the text vectorization feature, you must select the analyzer that is configured with "Vector - K-12 Education Vector".
Multimodal retrieval
The OpenSearch Industry Algorithm Edition for education automatically generates a multimodal search policy named sys_strategy for your instance. To test the multimodal retrieval feature, you need to configure specific parameters on the Feature Extensions > Search Test page. The required parameters are as follows:
Required parameters
-
query=default:'Education'
-
qp=sys_default
-
raw_query=Education
-
search_strategy=sys_strategy
Notes
-
You can upgrade a General-purpose application to an Industry Algorithm Edition application, but you cannot downgrade an Industry Algorithm Edition application to a General-purpose application.
-
The "Vector - K-12 Education Vector" index cannot be used alone. It takes effect only when the text vectorization feature for query analysis is configured on an index that uses the "Industry - Education Question Search" analyzer.
-
The OpenSearch Industry Algorithm Edition for education is applicable only to applications on an exclusive cluster.
-
If you want the system to automatically generate vectors for a field, you must set the field type to TEXT when you add the field. When you configure the index for the field, you must set the analyzer to "Vector - General Education Vector" or "Vector - K-12 Education Vector".
-
In the application schema, you must assign the required field tags to the fields. Otherwise, an error occurs.
If the field tags are not set correctly, a dialog box appears with the message "Please select the necessary field tags." You must click OK and return to add the required tags.
-
In the index schema, you must assign the required index tags to the indexes. Otherwise, you cannot proceed to the next step.
For example, if the index tag for the vector index is left as the default "Please select", the Next button is disabled. You must select the correct index tag before you can continue.