Build machine learning models in minutes, without a single line of code
Build, train, and use preset machine learning models in five proven steps
1
Identify business challenge
2
Gather applicable data
3
Generate a model
4
Evaluate results and learn
5
Choose how to use your model
Deploy to production
Generate insights
Automated Data Science Workflows in Kepler.
The AI and Machine Learning workflows within Kepler allow businesses to scale AI across a wide range of business use cases that cross all industries and functions, from forecasting demand and workforce planning to medical image analysis and anomaly detection.
Kepler can be applied to hundreds of use cases across your organization.
Tabular Classification
Predicts which category structured data will fall into.
Examples of use cases:
- Customer churn
- User intent prediction
- Loan application prediction
Tabular Regression
Measures the relationship between dependent and independent variables.
Examples of use cases:
- Customer lifetime value prediction
- Demand forecasting
- Contract duration analysis and prediction
Text Classification
Predicts the category of unstructured text data.
Examples of use cases:
- Sentiment analysis
- Topic classification
- Content compliance
Time Series Forecasting
Predicts a future quantity based on historical, sequential data.
Examples of use cases:
- Demand forecasting
- Raw materials price movements
- Trend prediction
Anomaly Detection
Finds outliers in data that differ significantly from the majority of the data.
Examples of use cases:
- Sensor data monitoring
- Fraud detection
- Anomalous behavior detection
Clustering
Clusters groups of unlabeled data into categories according to patterns and similarities.
Examples of use cases:
- Customer segmentation
- Personalized product bundling
- Precision marketing
Image Classification
Categorizes images that share characteristics of a predefined set of categories.
Examples of use cases:
- Image categorization
- Medical imaging analysis
- Quality assurance
Recommender
Evaluates similarities between a user’s purchase behavior and the purchase behaviors of other users in order to make highly relevant recommendations.
Examples of use cases:
- Product recommendations
- Content recommendations
Explore the different ways you can use Kepler to gain agility in your business.
The Kepler platform’s Automated Data Science Workflows: How they work.
How Kepler makes the development of machine learning models easier:
You upload your data to Kepler.
You select the right Automated Data Science Workflow based on your business goal and desired outcome.
Kepler organizes and visualizes your data.
Kepler provides an in-depth view of your data to ensure it suits the needs of your project.
You give Kepler important context about your data, model preferences, and budget.
Kepler makes your data sparkling by removing blanks, standardizing texts & removing special characters.
Kepler tests, creates and improves features to build your ML model.
Kepler selects algorithms, undergoes automated configuration and hyperparameter optimization.
Kepler interprets results via dashboard, showing specific, clear details on data points within the machine learning model.
Kepler generates predictions from either static or real-time data.
Kepler generates code to connect inference to your system of choice, so you can start seeing results from your work.
Not all your data is the same.
The Kepler platform leverages a wide range of data types, allowing you to use more of the business’ data to generate insights, create forecasts, and make better decisions.
Tabular
Workflows
Text
Workflows
Image
Workflows
What data systems does Kepler connect to?
- Local and Network Drives
- Microsoft Azure Blob Storage
- Amazon S3
- MySQL
- MS SQL
- PostgreSQL
- Microsoft Azure SQL