Skip to content

What is a Dataset?

A dataset is a collection of papers stored in one folder on your My Content Dashboard. You can sort your dataset collection in descending order, from newest to oldest.

What is a dataset.png

Read Transcript
This is your Workspace.

After you log in, you will see your Dashboard. Here is where all your content will be stored and this is also a starting place for using any tools.

To get started, you need some documents to work with. To create a new dataset, you can click one of the “Add Content” buttons.

There are three ways to create a new document set: You can choose to proceed with the default integrated Research Papers collection, which is pre-selected to streamline the process and save you time when creating your dataset. You can share a link to a Google Drive or Dropbox folder containing files in the following formats: .pdf, .doc, .docx, .pptx, .txt, .html, .xlsx, or .xls. You can upload a document collection in the following formats: .csv, .bib, .nbib, .ris, or .xml.

Finally, you can create new datasets by using the different tools within Iris.ai - for example, searching in an Research Papers dataset with the Explore tool will create a new dataset.

You can view the content of a Dataset by either double-clicking the Dataset or choosing the “preview” option from the right-click or three-dot menu. Here you will see information about the data set including details on how the dataset was created.

If your dataset has more than 10 thousand documents, you will be asked to use filters to narrow it down to be able to load the documents.

By clicking on the document from the list, you can see the title and the abstract and the link to full text, if available. Each dataset is the starting point for using any other tools. From there you can Explore, Analyze, Summarize, Filter, Extract or Chat with documents to find exactly the information you need.