HERE WE GO! New features just launched in Relevance AI 🚀🚀🚀
✅ A new document view feature 📄
✅ Upgraded AI Tagging experience 🏷️
Let's take a look inside! 👇
Our new document view provides one place to move and edit your documents, as well as find similar data / responses in your dataset.
While our AI workflows eliminate most of the work when it comes to structuring your unstructured data, there are always tweaks users want to make. For example, they may want to move a document from one category to another. We’ve focused on making this experience really easy!
When viewing a document anywhere in Relevance AI, hovering over it will reveal these options:
Click any of them to open up the new document view.
The left icon takes you straight to the “Move to” page, the middle icon takes you straight to the “Find similar documents” page and the right icon lets you view the document in full.
When viewing the document in full, you’re able to see all fields in the document (beyond the ones you are previewing in your dashboard view).
You can also click the “Edit” button in the top right hand corner to tweak editable fields and save your changes.
Click “Move to” to move the document from one category (cluster) to another. You’ll see a list of all the cluster options on the right. Click the one you want to move it to.
You’re able to select multiple documents at a time to move by clicking “Explore all data” in your Category View page and selecting multiple checkboxes.
All selected documents will now appear in the left column on your “Move to” screen.
Finally, click the “Find similar documents” button to use AI semantic search to find similar data in your dataset. To enable this feature, you will have to have run a Vectorize workflow on your text field (automatically done for you in AI Clustering flow).
Taking on feedback from you, we’ve improved our AI Tagging experience to make it easier to get started.
Our AI Tagging flow makes it super intuitive to use AI to tag your data, saving you days of manual coding. However, feedback suggested it wasn’t clear what to do to get started. To address this, we’ve improved the user experience of launching your AI Tagging workflow.
We’ve condensed the first step of AI Tagging into a simple step-by-step wizard, with detailed explanations. It takes only three decisions to launch your workflow!
You’re able to create your own list of tags - or paste in a comma separated list of tags that you already have handy.
However, we provide two utilities to “Suggest tags” based on the contents of your data. The first is “Most common words”, which will analyse your data and create tags from the most commonly used keywords. This happens instantly.
You can see below, tags have been added to my list. I’m able to remove ones I don’t like. For example, there are a lot of reviews about Jim Beam products in this dataset - so it has identified “Jim” and “Beam” as common words. However, these don’t make for good tags. This demonstrates a limitation of this strategy.
If I want more advanced and intelligent tag recommendations, we can harness our second option: AI suggestions.
Rather than simply find the most common words, this will use AI to analyse your data and determine a list of good tags. This may find tags that you wouldn’t have thought of, and don’t exist strictly as words in your data. However, this can take up to an hour depending on the size of your dataset.
When you have your list of tags ready, you can select the maximum amount of tags you want our AI to apply to each document.
Then click Get Started and you’re off to the races!
You’ll receive an email when this is complete! 📧
Your data will be tagged with your list of tags, thanks to the power of Relevance AI.
✅ We have greatly improved our AI category labelling algorithm in this release. You should see better category labels generated! Re-generate if you want to try.
✅ You may have noticed our “Delete All Data” button was broken. It is fixed!
✅ You can now add sub-headers to your Deployable Group menus.
Updated 3 months ago