The most exciting reason (in my opinion) for upgrading from the free version of Watson to Watson Professional is the additional data sources. In prior chapters, all of our sample use cases were based on CSV files. With Professional, we have many more choices to explore.
From the Welcome page (the free version), if you click on Add and then on Upload Data, you will see this:
In the Professional version, you see the following screen (note that Browse for locating and uploading a CSV file is still supported, but it is not shown in my screenshot here):
Under Select a source, you have the following options for loading of data: Browse (also available in the free version), Twitter (also available in the free version), IBM Cognos Report, DropBox, Box, and OneDrive (all new options).
In Watson professional, you have the option to create datasets based on data that is contained in on-premises and in-cloud databases. With Watson administrator access, you can set up and manage external data connections so that you can access this type of data with Watson. From the Where is the data you want to upload? page, using the section named Or select a connection created for you, you can search for and sort the list of available connections by their type, name, description, and last modified date, as shown here:
Since we've just upgraded (to Professional), this page displays No connections available. To create a data connection, you need to be a Watson administrator for your account. From the Watson main page, click on the account icon, which is shown here:
Then click on Administer. You'll see that the administrator accounts are indicated in the user list with the crown icons to the left of the username, as shown in the following screenshot:
This is the non-administrator view:
The following is the administrator view:
To create a data connection, an administrator can click on Data connections (shown in the next screenshot) and then on Create new connection:
In the Select a connector type dialog, you can click on the data connection that you are interested in setting up, as follows:
Let's take a look at setting up a data connection using Microsoft SQL server as an example. To create such a connection, you need to provide specific configuration information. From the Select a connector type dialog (shown in the preceding screenshot), you can click on the Microsoft SQL Server icon. Then, you can use the Watson Analytics administration console, as shown in the following screenshot, to set the required properties:
To keep things simple, the only required settings that you will need to provide are the following:
- Connection name: This is just an identifier for the connection within Watson
- Server name: This is the actual computer name or IP address of the computer on which SQL server is running
- Port number: This is the port number on which SQL server is running
- Database name: This is the name of the SQL server database that you wish to access
Once your connection is set up, it will appear as an option for uploading data under the Add tab on the Watson main page, as follows:
When you click on the data connection, Watson will prompt you for your user ID and password, like this:
Upon successful connection (to your database), Watson provides the Import from data connection dialog (shown in the next screenshot), where you can:
- Type a SQL statement
- Run the SQL statement and preview the results
- Upload the SQL results to Watson:
Once your data is uploaded to Watson, you can explore, predict, or assemble it (and do much more) just like any other data file you have already uploaded!