Sign Up & Get Onboard
High-level Introduction to Relevance AI
Relevance AI's onboarding process includes two main steps.
- Sign up where you create your account on the platform
- Analysing your first dataset through the onboarding wizard
Below is the full description of the process.
1. Sign up
- Go to relevanceai.com and click on Sign up.
- Enter your email address and create a new password and click on Continue.
- You will be directed to a new page to set up your profile. Enter your name, company name, role and set your closest data region (based on your current location).
Now that you have a profile, you're ready to analyze your first dataset!
2. Onboarding wizard
Through this wizard:
- You can enjoy the provided demo for some of the frequently used analysis available on Relevance
- Ask Relevance is to search through documents (such as PDF)
- Cluster AI groups your data entries based on their semantical similarities
- Interview AI extracts key information out of interview files (e.g. audio transcription)
- Document AI extract key information out of text data
- You can also upload your one dataset (we recommend reading our guide on Dataset Requirements in advance) and chose your desired workflow (i.e. analysis).
Upload my own data
- Choose your data type. You can also use the provided sample datasets.
- Follow the prompts and you see a few entries of your dataset. Using "Remove columns", you can remove some of the fields in your data.
### Analyse the data / Run an AI workflow
Clicking on any of the Demos will open their corresponding workflow setup wizard page. If you selected to upload your own data earlier, after upload is completed, you will be directed to the general Workflow page where you chose what analysis you wish to apply to your data.
Frequently used workflows are listed on top. For text data, we recommend AI Text clustering, AI Tagging, Extract Sentiment and Enable Search.
Read through the names and descriptions to select the suitable workflow. You can also search based on the workflow use cases, functionality, etc.
Explorer: Your dashboard to insights
All analysis results are written back to your dataset and can be viewed for analysis on the Explorer. You can use auto-generated dashboards (one way of accessing it is shown in the image below) or build your dashboard from scratch.
A sample dashboard on clustering results is explained below:
AI describes your top themes automatically
Here you'll be automatically prompted to use AI to label your top themes.
To continue, click on 'Use AI to label your categories'.
At any stage, you are encouraged to manually label some of the categories to personalise how you would describe a category. Next to the category name, click on the 'pencil' icon and type in how you would label / describe the shown items / responses within that group.
Then click on 'Autolabel AI Categories' to update the labels for all categories. The underlying model can learn your preferred structure for description by seeing a few examples on how you describe a group. This is why you are encouraged to enter a description for two (or so) of the groups/themes/trends.
RECAP - Want to see all of the steps in action? (one min)
Check it out!
Find insights through clustering
Updated 3 days ago