Sentiment Analysis Example

Use sentiment analysis to determine if a phrase is positive or negative. For example, you may want to see how many customer reviews were positive or negative.

In the following example, a table containing amazon reviews of a book is used as the datasource.

The input must be a text column where each line contains more than one words.

Add the Text file source to the canvas, and add the upload the file containing the text column.

Set the properties for the Text file source, ensuring that the delimiter is not a space, and the add the node.

The table preview should contain a column, with a sentence in each row.

Drag the Marketplace node onto the canvas, click on it, and download the Sentiment Analysis algorithm from the dialog.

Connect the Marketplace node to the table node, and in the Properties panel, define the source input and output. Under Output options, define whether to add the new column to the existing table, or create a new table.

Note that by default, the name of the new column is set to classifiedColumn.

To change this, click the output row and enter the required column name. Then go to line 26 of the script and replace classifiedColumn with the new column name.