Ask data questions

Prerequisites

1. Create a dataset and configure Intelligent Data Assistant Q.

2. Obtain the necessary resource permissions. For more information, see permission management.

PC entry points

• Entry 1: On the Quick BI homepage, click Intelligent Assistant Q, and in the main input field, select Intelligent Data Assistant Q.

• Entry 2: In the left-side navigation pane of the Intelligent Assistant Q homepage, select Intelligent Data Assistant Q.

Data sources

Datasets

You can query datasets under the All tab and within each analysis topic for which you have permissions.

• To configure dataset permissions for Intelligent Data Assistant Q, see Manage permissions for Intelligent Data Assistant Q.

• To configure an analysis topic, see Manage analysis topics.

• Datasets marked with a Sample tag in the upper-left corner are sample datasets.

  

  If you do not need sample data, you can hide it. For more information, see Display sample datasets.

Data files

You can upload local data files.

Upload a data file

1. On the Data Files tab, if no files have been uploaded, click Upload File.

   If you have already uploaded data files, click the Upload File icon in the upper-right corner to go to the upload page.

   

2. On the file upload page, drag a file to the upload area, or click the area to select a file.

   

3. Preview the data.

   • On the data preview page, you can set the display name for the dataset.

     

     Note

     Enter a valid name. Do not use special characters or leading/trailing spaces.

   • You can perform a data preview and view field details.

     

   • After confirming the data is correct, click Save and Ask or Save Only.

     

     • Save and Ask: Saves the file as a dataset and opens it for querying.

       

     • Save Only: Saves the file as a dataset without opening it for querying.

Once the file is uploaded, you can select it to ask questions.

Manage data files

• Configure

  • Hover over a data file, click the icon, and select Configure.

  • On the configuration page, you can modify the display name, use data preview, and view field details.

• Delete

  • Hover over a data file, click the icon, and select Delete.

    

  • In the confirmation dialog, click OK to delete the data file.

    

Preview and select datasets

Select a single dataset

• In the Intelligent Data Assistant Q interface, click to select a data source. You can see the datasets and uploaded data files for which you have permissions. Hover over a dataset to Preview it or click Ask to query it directly.

  

  • Click Preview (①) to view the dataset's field details and data preview. From here, you can use quick questions or click Ask in the upper-right corner to open the Intelligent Data Assistant Q chat box.

    • Field Details

      

    • Data Preview

      

      Hover over a dimension field to view its details.

      

    • Quick Questions

      

    • Ask

      Start asking questions about the current dataset.

      

  • Click Ask (②) to open the Q&A interface for the current dataset.

• After selecting a dataset, you can click the icon next to its name to preview its data fields, including key metrics and analysis dimensions.

  

  Click the icon again to collapse the field view.

  

• During your conversation, you can preview data and switch datasets.

  • Preview data

    Click the icon next to the selected dataset name to open the data preview.

    

  • Switch datasets

    1. Click the name of the selected dataset to view available datasets.

       

    2. Hover over a target dataset and click Ask to switch to it.

       

       You can also Preview the target dataset before clicking Ask.

       

Select multiple datasets

If your question requires data from multiple datasets, you can select and query them together.

1. Enter your question.

   The system displays potentially relevant datasets. You can select the ones you need.

   

2. Select the checkboxes for multiple datasets and click Ask based on the selected data.

   

3. View the analysis result.

   

Ask questions and get answers

Enter a question in the chat box and send it. You can type questions directly, use quick questions, or use voice input.

Enter a question directly

• For example, if you enter "What is the proportion of orders for each order level?" and send it, the system returns a data analysis result and displays the analysis process on the right.

  If you do not need to see the analysis process, click the icon on the right to hide it.

  

  When hidden, only the chart appears.

  

  You can click Analysis Complete to expand the analysis process again.

  

• As you type, the system may suggest recommended fields and recommended questions. For example, typing "2024" might display the following suggestions, which you can select based on your needs.

  

Inference and interpretation modes

Overview

For both inference and interpretation, you can choose between a built-in large model and a custom model. Select the modes that best fit your analysis needs.

Examples

The following examples show the results of asking the question, "What is the proportion of orders for each order level?" using different inference and interpretation modes.

• Example 1: Use Qwen3.7-plus for inference and no data interpretation.

  The system uses the Qwen3.7-plus model to return the inference process and a chart.

  

• Example 2: Use the built-in Qwen3-max model for both inference and data interpretation.

  The system uses the Qwen3-max model to return the inference process, data interpretation, and a chart.

  

• Example 3: Use the built-in DeepSeek-R1-0528 model for both inference and data interpretation.

  The system uses the DeepSeek-R1-0528 model to return the inference process, data interpretation, and a chart.

  

Intelligent planning mode

With intelligent planning enabled, Intelligent Data Assistant Q can handle multi-step complex questions and multi-turn conversations.

• Enable intelligent planning mode to answer a multi-step complex question.

  Ask a question like, "How were the sales in East China in 2024? Provide sales suggestions based on recent data trends." The system integrates the data results and provides recommendations.

  

• Enable intelligent planning mode to have a multi-turn conversation.

  First, ask, "Which province had the highest sales in 2024?". After getting the result, switch datasets and then ask, "What is the top-selling product in this province?". The system automatically carries over the context from the previous turn and provides an answer.

  After you obtain the query result, switch the dataset and enter "What is the name of the best-selling product in this province?". The system will then automatically carry over the context from the previous conversation and return a result.

  

Ask a follow-up question

Follow-up questions let you drill deeper into a previous result for more precise data exploration.

1. In the conversation interface, click Follow-up Question to ask for more details about the previous question, "Which province has the highest sales?"

2. Enter your follow-up question based on your needs, for example, "What is the top-selling product?".

   Click the icon in the lower-right corner of the chat box or press Enter to send the question and get the analysis result.

   Note

   The system automatically exits the follow-up question mode. If you want to ask another follow-up question, you must click the Follow-up Question button again.

Conversation list

• Click the icon in the upper-left corner to view your conversation history.

  

• Click the icon to start a new conversation with Intelligent Data Assistant Q.

  

Mobile client

• Mobile micro app configuration

  Follow the steps shown in the image to enter the mobile editing interface.

  In the editing interface, configure Intelligent Data Assistant Q for the mobile micro app.

  

• Result

  Intelligent Data Assistant Q is available on the mobile client. The operations are similar to those on PC.

  

Tips for asking questions

1. Specify a time frame in your questions. For example:

   1. Instead of "sales proportion by product", try "sales proportion by product this year".

2. To avoid ambiguity when aggregating multiple dimension values, use keywords like "each", "per", "total", or "combined". For example:

   1. A question like "sales for Zhejiang and Jiangsu" or "sales from 2020 to 2023" can be ambiguous. Rephrase it as "sales for Zhejiang and Jiangsu each", "sales per year from 2020 to 2023", "sales for each year from 2020 to 2023", "total combined sales for Zhejiang and Jiangsu", or "total sales for Zhejiang and Jiangsu".

3. For questions about maximum or minimum values across multiple dimensions, specify whether you want grouped or ungrouped results. For example:

   1. Instead of "What is the best-selling product in Zhejiang and Jiangsu?", try "What is the top-selling product by sales in Zhejiang?", "What is the top-selling product by sales in Jiangsu?", or "What is the top-selling product by sales for each province?".

   2. Instead of "What were the 3 best-selling months last year for Zhejiang and Jiangsu?", try "What were the 3 best-selling months last year for Zhejiang?", "What were the 3 best-selling months last year for Jiangsu?", or "What were the top 3 order months by sales for each province?".

4. If a dimension's value could be ambiguous, specify the dimension name clearly. For example:

   1. A question like "Rank products purchased by corporate customers by sales" can be ambiguous because "corporate" might be misinterpreted. Be more specific: "For products where the customer type is corporate, rank them by sales."