Introduction to One AI

One AI enables HR professionals to access statistical and machine learning insights alongside their people analytics data in One Model. It empowers informed and ethical people decisions because it is the only fully-explainable machine learning platform designed specifically for HR.

Core Capabilities

One AI delivers the following core capabilities, all from within One Model:

Machine Learning

One AI provides a full auto-ML platform producing predictive and causal models on your people data. While many use cases exist, the most common is predicting attrition risk within your employee population.

  • Classification and Regression models can be created

  • These models are trained on YOUR data

  • One AI provides transparency into model configuration and performance

  • A guided data framing utility (One AI Recipes) enables individuals to create and understand models without requiring them to be Data Scientists

  • Results can be deployed into your data model and viewed in Storyboards

To learn more, please review the Introduction to Machine Learning in One AI article.

Embedded Insights

These services integrate statistical capability directly into Storyboard tiles. If data is detected in a format where statistics can be applied, a lightbulb icon appears allowing you to enable the statistical overlay.

  • Trend forecasting with advanced capabilities such as seasonality detection

  • Correlations between factors in scatterplots with statistical significance explanations and line of best fit

  • Table Insights that highlight potentially noteworthy or unusual areas in your company by leveraging descriptive statistics to add information and visual cues to Storyboard tables.

Data augmentations

Your HR data can be augmented with commute time and distance based on zip codes that already exist in your data.

Ethics, Transparency, and Interpretability

The tagline for One AI is “Ethical and Transparent AI in HR”. One AI was created with these principles in mind and they have remained a focus throughout the product’s existence.

One AI enables transparent and ethical AI

  • Your data does not leave your One Model instance

  • Models are trained on your own data

  • Model performance data is readily available for every type of model (precision, recall, etc.)

  • Estimator selection and configuration are shared

  • Prediction explanations are provided

AI is more useful when it’s interpretable

  • Guided data framing is available for specific HR topics in our Recipes feature

  • Detailed Exploratory Data Analysis (EDA) reporting is front and center

    • This provides transparency into the data cleaning process

  • SHAP explanations are provided to explain classification models

  • Storyboard content offerings are available from the model outputs

Was this article helpful?

2 out of 3 found this helpful

Comments

0 comments

Please sign in to leave a comment.