Quick Guide to Find and Create Insights

Welcome to Relevance AI! To help you get started, check out the video below for a quick tour, including a demo on how to upload your first dataset.

Step 1: Upload Data in Relevance AI

You must be an admin or an editor to upload datasets. You can upload datasets with in Relevance AI via CSV file upload, via a Rest API using our Python SDK or via our Snowflake connector (coming soon!).

Dataset Requirements: Text

  • Make sure your file is a valid JSON or CSV.
  • Field names must contain letters, numbers, dashes or underscores. Please remove any spaces in the headings / field names (in contrast, spaces in the actual dataset are ok).
  • Have at least 1 string field to analyse text, filter and break down.
  • Have at least 1 numeric field to if you would like to create a measure (i.e calculate a metric such as average) in Relevance AI (optional).
  • For time series analysis, must be transactional or time-series data with at least 1 field configured as a date (“yyyy-mm-dd” format) or datetime data type (optional).
  • We recommend adding an "_id" column with the IDs of the documents. Otherwise, we will automatically create an ID for each row.
  • Maximum CSV file size— Theoretically, there is no limit to the amount of data that Relevance AI can handle. Your data is stored on the cloud that scales based on user demand, potentially reaching terabytes of stored data.

Which preprocessing techniques are recommended when working with text?

When working with textual data it is recommended (i.e. not required) to apply certain preprocessing steps which can potentially improve the analysis results. Common text pre-processing are:

  • Stop words removal: to remove frequent but not important words used in our language (e.g. the, there).
  • Stemming: replacing words with their word stem (e.g. changes or changing become chang-)
  • Lemmatization: replacing words with their common root (e.g. changes or changing become change)
  • Lowercasing: converting all characters to their lowercase form
  • Text cleaning: this step is completely data specific. Some famous text cleanings are Html, URL or hashtag removal.
  • Breaking into shorter pieces of text: when automatically analyzing text, processing smaller pieces of text (e.g. a sentence vs paragraph) often produces more precise results.

Dataset Requirements: Images

Coming soon

Dataset Requirements: Audio

Coming soon

Step 2: Follow The Prompts (Onboarding Flow)

Continue to follow the prompts to complete the onboarding flow.

Step 3: View Your Dashboard

Welcome to the Explorer page! The Explorer page is the best place to find and create in-depth insights about any areas within your dataset. Here you can view insights and trends at a high level, dive deep into the details, add charts, personalise views and more! Read more about Explorer.

Step 4: Add Additional Workflows

Enrich your data with additional Workflows. Read more about workflows at What are workflows.

Step 5: Publish and Share Dashboards

Select the dashboard/data you want to share and select Share. Read more about User Management.

TIP: Analyzing More Than 100MB?

Both the onboarding wizard and 'import CSV' are no-code ways of uploading data to get insights in minutes. Simply drag and drop your CSV file to upload data, follow the prompts and get insights. That said, the wizard is limited to analyzing CSV files no more than 100MB. To upload data files greater than 100MB, follow our Python workflow here.