DataWorks:Improve accuracy by using a knowledge base

Configure question templates

Go to the knowledge base page and click the Question Template tab, where you can add a new question template in two ways:

• Add a template from the UI.

  1. On the knowledge base page, click Add to open the Create Question Template page.

  2. Configure the parameters for the question template:

     Parameter	Description
     Question	Define a common question used in a session. For example: Monthly sales trend of Product B in Store A. The question can include variables in the ${parameter_name} format. If you define variables, you must provide enumerated or sample values in the Parameters field. For example: Monthly sales trend of ${commodityName} in ${shopName}.
     SQL	Provide an accurate and executable SQL query based on the question. If the question contains a ${parameter_name}, the SQL must also include a parameter with the same name. For example: SELECT shop_name, commodity_name, sales, month FROM ads_order_table WHERE shop_Name = ${shopName} AND commodity_Name = ${commodityName} ORDER BY month;
     Parameter	If you use variables like ${parameter_name} in the question or SQL, you must provide a list of possible or example values here. For example: ${parameter_name}: ["value1","value2","value3"]
     Datasets	The datasets this template applies to. If empty, it applies globally.

  3. After you complete the configuration, click Save.

• Bulk import templates from a file.

  1. On the knowledge base page, click Import and Add.

  2. In the import dialog box, click Download Template. In the template file, fill in the Question Template, Template Script, Template Parameters, and Datasets columns. The parameters are the same as those used for UI configuration.

  3. Drag and drop the local file, or click + in the dialog box to select the file to import.

  4. After you add the file, click Confirm.

Configure term management

Go to the knowledge base page and click the Term Management tab. You can add a term in one of two ways:

1. Add a term from the UI.

   1. Click Add to open the Create Term page.

   2. Configure the key parameters for the term:

      Parameter	Description
      Term	A common or specialized word used in business communication and analysis. For example: month-over-month.
      Alias	Synonyms, abbreviations, full names, or translations of the term. Press Enter after each alias to add multiple aliases.
      Datasets	Specifies the datasets this term applies to. If left empty, the term applies globally.

   3. After you complete the configuration, click Save.

2. Bulk import terms from a file.

   1. On the knowledge base page, click Import and Add.

   2. In the import dialog box, click Download Template. In the template file, fill in the Term, Alias, Description, and Datasets columns. The parameters are the same as those used for UI configuration.

   3. Drag and drop the local file, or click + in the dialog box to select the file to import.

   4. After you add the file, click Confirm.

Configure business logic

Go to the knowledge base page and click the Business Logic tab. You can add business logic in one of two ways:

1. Add business logic from the UI.

   1. Click Add to open the Create Business Logic page.

   2. Configure the key parameters for the business logic:

      Parameter	Description
      Business Concept	The name of a common concept in business communication and analysis. For example: valid order.
      Logic Explanation	A definition for the business concept. For example: ord_cnt_1d > 0 indicates a valid order.
      Datasets	Specifies the datasets this business logic applies to. If left empty, the logic applies globally.
      Field	Associate the business concept with specific table fields from the selected dataset.

   3. After you complete the configuration, click Save.

2. Bulk import business logic from a file.

   1. On the knowledge base page, click Import and Add.

   2. In the import dialog box, click Download Template. In the template file, fill in the Business Concept, Logic Explanation, Associated table fields, and Datasets columns. The parameters are the same as those used for UI configuration.

   3. Drag and drop the local file, or click + in the dialog box to select the file to import.

   4. After you add the file, click Confirm.

Edit a knowledge base object

1. In the left-side navigation pane, click Knowledge Base to go to the knowledge base page.

2. On the tab for the desired object type, click Operation in the Edit column.

   The parameters are the same as described in Configure the knowledge base.

Search for a knowledge base object

• To find an object, go to the relevant tab, enter a keyword in the search box, and press Enter.

• You can also click a knowledge base object to view its details.

Delete a knowledge base object

To remove an object, find it on the appropriate tab and click Operation in the Delete column.

Scenario

An e-commerce company uses ChatBI to analyze sales data. The underlying table schema is as follows:

• Table name: ads_order_table

• Key fields: shop_name (store name), product_name (product name), sales_amount (sales amount), order_date (order date), and valid_order_flag (valid order flag, where 1 indicates a valid order)

Case 1: Value of question templates

SELECT SUM(sales_amount)
FROM ads_order_table
WHERE shop_name LIKE '%East China%'
  AND product_name LIKE '%iPhone%'
  AND order_date >= '2026-02-01'
  AND order_date < '2026-03-01';

SELECT SUM(sales_amount) as total_sales
FROM ads_order_table
WHERE shop_name = ${shopName}
  AND product_category = ${productCategory}
  AND order_date >= ${startDate}
  AND order_date < ${endDate}
  AND valid_order_flag = 1;

${shopName}: ["East China Flagship Store", "South China Flagship Store", "North China Specialty Store"]
${productCategory}: ["iPhone", "iPad", "MacBook", "AirPods"]
${period}: ["last month", "this month", "last week", "this week"]

SELECT SUM(sales_amount) as total_sales
FROM ads_order_table
WHERE shop_name = 'East China Flagship Store'
  AND product_category = 'iPhone'
  AND order_date >= '2026-02-01'
  AND order_date < '2026-03-01'
  AND valid_order_flag = 1;

Case 2: Value of term management

WITH monthly_sales AS (
  SELECT
    DATE_FORMAT(order_date, '%Y-%m') as month,
    SUM(sales_amount) as sales
  FROM ads_order_table
  WHERE shop_name = 'East China Flagship Store'
    AND valid_order_flag = 1
    AND order_date >= '2026-01-01'
    AND order_date < '2026-03-01'
  GROUP BY DATE_FORMAT(order_date, '%Y-%m')
)
SELECT
  (curr.sales - prev.sales) / prev.sales * 100 as mom_growth_rate
FROM monthly_sales curr
JOIN monthly_sales prev
  ON prev.month = '2026-01'
  AND curr.month = '2026-02';

Case 3: Value of business logic

SELECT SUM(sales_amount)
FROM ads_order_table
WHERE shop_name = 'East China Flagship Store'
  AND order_date >= '2026-02-01'
  AND order_date < '2026-03-01';

SELECT SUM(sales_amount) as valid_sales
FROM ads_order_table
WHERE shop_name = 'East China Flagship Store'
  AND order_date >= '2026-02-01'
  AND order_date < '2026-03-01'
  AND valid_order_flag = 1
  AND sales_amount > 0;

Verifying the knowledge base

After configuring the knowledge base, verify that it works correctly:

1. Ask questions in a ChatBI session: Use the example questions from your configured question templates and check if the generated SQL meets your expectations.

2. Test term recognition: Use aliases of configured terms (such as "MoM") in your questions and observe whether ChatBI understands them correctly.

3. Verify business logic: Ask questions involving business concepts (such as "valid order") and check if the generated SQL includes the correct business logic filtering conditions.

4. Compare results: Record the change in question-answering accuracy before and after configuring the knowledge base to quantify the improvement.