Keywords from any text can help us understand what the text is about. With help of Bitskout, you can extract general keywords or even create your own custom keywords using examples.
You can even run both models at the same time. This can be useful in the following use cases:
extract information from a support ticket
classify & sort blog articles
extract data from CVs or press releases to search it later
To create a keyword extraction model:
Go to A.I. models and press Add to add a new model.
Specify the model name and choose the Text Creation option.
Scroll down to the list of options and choose Extract Keyword.
Once you choose the Extract Keywords option, you will see the setup screen:
The first thing we need to set up is a type of content. There are several types of content we support. If you are working with something that is not specified in the list, then please choose Text.
The next thing is the Creativity setting. If you put close to 1, then A.I. will become very inventive with keywords. We recommend starting at 0.3 - 0.4 and adjusting based on results.
The next is setting up the examples. If you set the examples, then A.I. will use them as a reference to create custom keywords. If there are no examples, then A.I. will extract general keywords.
Once you finished configuring the options, choose the option "Automatically create a workflow .." and press Apply.
Now the model is ready and you can proceed to create a custom output to write the data to your tool.