This offer allows you to deploy a fully functional version of Kanva in your Microsoft Azure environment, free of charge except for Azure resource consumption, for trial purposes.

Kanva: Empower Your Business with Predictive AI

Kanva is an innovative AutoML solution targeted at domain and subject matter experts who don't necessarily possess coding or data science skills. It combines the most powerful machine learning models for tabular data with the most user-friendly interface available in the market. With Kanva, your domain experts can create models to predict customer churn, forecast sales, predict risk of incidents, implement predictive maintenance schemes, and address a plethora of other use cases.

Key Benefits

Accelerate AI Adoption
Adopting AI is hard: 70-90% of companies' AI initiatives fail due to high costs, skill gaps, and lack of domain expert involvement, Kanva makes it easier by lowering implementation costs, removing the need for technical skills, and engaging domain experts from the outset. This approach significantly reduces risk and increases ROI.
Unlock Distributed Innovation
Kanva unlocks distributed innovation by providing an accessible AI tool that enables individuals across an organization to independently create predictive models, optimize workflows, and make data-driven decisions.
Engage Domain Experts Early On
Leverage their expertise to identify opportunities where AI can drive tangible business value.
Future-proof
Kanva ensures future-proofing for organizations by offering continuous capability and algorithm updates, seamless project handovers, and easy integration with data science teams.
Enhanced Security and Privacy
Kanva operates entirely within your infrastructure, ensuring that you maintain complete control over your data and predictive models.

Features

Supported Model Types

Forecasting
Binary Classification
Multiple Classification
Regression

Model Interpretability:

For each individual prediction, overview of how much each driver influenced it and in which way.
Identify the most important/predictive drivers at a global level.

Data Insights:

Trends and Outliers
Correlation Analysis
Statistical Analysis

Panel Forecast: Group a dataset by a column, and create a forecast for each subset of data belonging to a group.
Hierarchical Forecasts: Similar to panel forecast, but a global forecast model is also created and reconciled.
Grouped models: Group a dataset by a column, and create a classification or regression model for each subset of data belonging to a group.
Supported Data Sources:

ODBC Databases
Web APIs
Files (Excel spreadsheets, CSV files, TSV files, parquet files)
Cognite Data Fusion

Export Models:

As Web APIs
As Binaries
As Jupyter Notebooks

Coming soon

Dimensionality Reduction: Express high-dimensional data in fewer dimensions while preserving key information, making it easier to visualize and analyze complex datasets.
Clustering: Automatically group data into clusters by similarity.
Scheduled model re-training
Scheduled data update
Scheduled batch predictions

Low Entry Barrier

You don't need big upfront investments to test Kanva. You can start with a free trial and switch to a permanent subscription when it proves to be a valuable addition to your organization.

Free Trial

Kanva installed on your infrastructure.
Create as many predictive models as you want.
No subscription fee.

Optional support in the form of a Pilot Project, billed as consultancy (approx. 45 hours at 850 NOK/h).
Maximum duration: 3 months.

Permanent Subscription

Kanva installed on your infrastructure.
Receive regular maintenance and upgrades, it works like self-hosted SaaS.
Billed as subscription with a base fee per installation + extra per user bracket (1-5, 6-10, etc.)
No consultancy fee (except for custom work).

Contact

You can contact Human-Driven AI, the Norwegian startup behind Kanva, for support or to discuss any opportunities at contact@human-driven.ai.

Kanva (Free Trial)Human-Driven AI

Kanva (Free Trial)

Human-Driven AI

Kanva (Free Trial)

Human-Driven AI

Kanva (Free Trial)
Human-Driven AI