January 2024 marked the release of 1.12.2: Improved data analysis, our first Generative-AI powered features and coupon codes.
In our first release for 2024, we are pleased to announce the following changes to the AI & Analytics Engine:
-
Improved Data Analysis: Improved column-selection UI, bigger charts, clearer statistical summaries, and two new functions to improve the overall usability
-
Our first Generative-AI powered features: We have taken our first step towards improving the automation and usability of the Engine, by leveraging Large Language Models and Generative AI. The Engine now provides improved suggestions for feature selection and feature-type configuration
-
Coupon codes: You can now receive coupon codes from us as part of our promotional offers, and apply them to your Organization subscription to get a discount
Improved Data Analysis page
The data analysis section of the dataset details page has been refreshed with an improved user interface.
Whether you are looking at the default automatically-generated analysis results (with no grouping) or your on-demand analysis results (with grouping), the “Single column” view now shows a larger visualization for your chosen column. The content shown in visualization tooltips has been made more prominent, and a search bar has been included to search for particular columns in large datasets.
The statistical summary accompanying each visualization in this view is now clearer, with improved number formatting and ordering of stats. If a visualization is not available due to the nature of the data in the chosen column, there is a clear indicator that visualizations are available for categorical and numeric data.
In the pairs view, we have made it easier for you to change the selection of columns:
Our first Generative-AI powered features
We are excited to introduce our first two Generative-AI powered features!
In the feature selection stage of regression or classification app setup, while asking the Engine to select features automatically, the Engine uses Large Language Models to understand the context of the dataset columns, allowing it to suggest which columns can be excluded from consideration:
In the manual feature-selection option, the same function is available while requesting feature importance score estimates to help with your selection:
We have also leveraged LLM and Generative AI to improve the Engine’s recommendation of feature types for regression, classification and clustering apps:
Coupon codes
You can now apply a coupon code when you sign up or upgrade your subscription on the checkout page, or apply it to an existing subscription on the organization’s billing page, where it will be applied for the next billing cycle. We’ll be offering promotional discount codes, so keep an eye out!