The do-it-yourself application to
create, train and deploy
your own machine learning models.
A No-Code AI interface for citizen data scientists!
The AI Studio allows business analysts, data analysts, and data scientists to rapidly develop Artificial Intelligence and machine learning models through a user-friendly and visual interface. Structured in four easy steps, the AI Studio supports the full AI lifecycle.
Craft training data
The toughest and most often overlooked part of any ML model is good training data. The Squirro AI Studio makes this process a breeze using Squirro’s Cognitive Search to define a Candidate Set and a "what you see is what you get" interface to quickly label data.
Create a model
Take a labelled data set - we call it a Ground Truth - and select from various algorithmic methods - including Proximity Search, Naive Bayes, SVM, Random Forest and Deep Learning. Meld the two and create your ML model.
Validate your work
A model is only as good as its accuracy. The AI Studio includes a comprehensive test suite to validate your model’s performance and allows you to work in operational feedback from the actual usage of your model.
Choose from the models you built or from our catalogue of pre-trained models and deploy them in Squirro’s Machine Learning Service. The result: Actionable Insights.
Automatically integrated with the Squirro Insights Engine
Continuous support from Squirro at all stages
Do-It-Yourself model building
Augmented Intelligence solutions and Insights Engine
Turn unused data into actionable
A powerful platform empowering data
analysts anda data scientists. That’s the
Squirro Insights Engine
- Craft your own labelled Ground Truth sets
- Build your own models based on your data
- Additional pre-trained models available out of the box
- Fully validated ML models
- Deployed in Squirro or exported for your use elsewhere
- Use our API based classification service with your ML models
- Full integration with Squirro Insights Engine
- Accessible to business users, not just data scientists
- Speeds up ML model creation by 80%
- Delivers deep insight at speed and scale
- Saves money with less need for data scientists
- Works with an organization’s own training data
- Users have the ability to train their own models