Skip to content

Commit

Permalink
update docs
Browse files Browse the repository at this point in the history
  • Loading branch information
lijing-22 committed Dec 19, 2024
1 parent e9d93f8 commit 7647e45
Show file tree
Hide file tree
Showing 8 changed files with 26 additions and 13 deletions.
39 changes: 26 additions & 13 deletions pages/docs/ai-chat/ai-data-collection.mdx
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# AI data collection

One of the highlights of Chat2DB's intelligent question-and-answer feature is its powerful ability to create custom AI datasets.
One of the highlights of Chat2DB's intelligent question-and-answer feature is its powerful ability to create custom AI collection.

With this feature, users have the flexibility to build different sets of Q&A data, so that the content of the Q&A is highly customized, ensuring that each answer is precisely tailored to your specific needs.
Not only that, this process also greatly improves the intelligence level of the question and answer system and the accuracy of the answer, so that you are more comfortable in dealing with complex and changeable query scenarios.
Expand All @@ -9,28 +9,41 @@ In short, the custom AI data collection is the key to unlocking Chat2DB intellig

## Step 1: Create an AI data collection

You can use each data source to create ai data sets for that data source. Tables from one or more different databases can be placed in an ai dataset.
You can also right-click all tables on the Tables node under each database to add a specified table to the ai dataset.
You can use each data source to create ai data sets for that data source. Tables from one or more different databases can be placed in an ai collection.
You can also right-click all tables on the Tables node under each database to add a specified table to the ai collection.

If your table structure changes after you synchronize, you can right-click the dataset node to resynchronize.
If your table structure changes after you synchronize, you can right-click the collection node to resynchronize.

!['sql-copilot'](/image/AI-image/create-ai-data-collection.gif)
1. Create an AI data collection
!['sql-copilot'](/image/AI-image/create-ai-data-collection1.png)

## Step 2: Modify the tag name and table comment
2. Enter the name of AI data collection
!['sql-copilot'](/image/AI-image/create-ai-data-collection2.png)

You can edit the table by double-clicking it within the ai dataset. If your business cannot modify the table name or table comment, you can modify the table name and comment as well as the field name and comment information here, so that AI can better understand your table structure and field information.
This will greatly improve the accuracy of ai generation.
3. Choose the database and select the tables you need.
!['sql-copilot'](/image/AI-image/create-ai-data-collection3.png)

!['sql-copilot'](/image/AI-image/ai-data-collection-edit-table.gif)
4. Sync to Al data collection
!['sql-copilot'](/image/AI-image/create-ai-data-collection4.png)

It will take a few minutes to finish, you can close this window.

## Step 2: Modify the AI Column Name and AI Column Comment

You can edit the table by double-clicking it within the ai collection. You can not modify your origin column name and comment here, you can modify the **AI column name and comment** as well as the AI Table Name and AI Table Comment, so that AI can better understand your table structure and column information.

This will greatly improve the accuracy of AI generation.

!['sql-copilot'](/image/AI-image/create-ai-data-collection5.png)

## Step 3: How to use the data collection

Wherever you call ai, you can select AI data sets to do more precise content generation
After set up the AI data collection, wherever you call AI, you can select AI data collections to do more precise content generation.

### Demo 1: Natural language generation SQL

!['sql-copilot'](/image/AI-image/ai-data-collection-text2sql.gif)
!['sql-copilot'](/image/AI-image/text2sql1.png)

### Demo 2: Generate reports in natural language
### Demo 2: Generate Dashboard in natural language

!['sql-copilot'](/image/AI-image/ai-data-collection-create-dashboard.gif)
!['sql-copilot'](/image/AI-image/dashboard.png)
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added public/image/AI-image/dashboard.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added public/image/AI-image/text2sql1.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

0 comments on commit 7647e45

Please sign in to comment.