RapidMiner takes important steps towards responsible artificial intelligence by helping analytical teams accelerate real-time to the value of IoT use cases.
Visual workflow designer
Increase the productivity of your computer science team, from analysts to experts
Fast and automate predictive modeling in the visual drag-and-drop interface
A rich library with over 1500 algorithms and functions ensures the best model for every use case
Predefined templates for common use cases, including customer outflow, maintenance forecasts, fraud detection and more.
Audience Wisdom offers proactive recommendations at every step to help beginners.
Connect to any data source
Work with all your data, no matter where it is
Create direct + click links with databases, corporate data warehouses, data lakes, cloud storage, business applications and social media
Easily reuse links at any time and share them with anyone who needs access
Connect to new resources with the RapidMiner Marketplace extension
Automated database management
Run data preparation and ETL in databases to optimize your data for advanced analysis
Query and retrieve data without writing complex SQL
Harness the power of highly scalable database clusters
Supports MySQL, PostgreSQL and Google BigQuery
News and bug fixes
New features
– Bias limitation – Receive bias notifications on all parts of the RapidMiner platform, including Turbo Prep, Model Simulator and more. When Studio thinks you have a column that could lead to model bias, you will receive a warning along with a platform call explaining why it was activated.
IIOT Advanced Streaming-Mix and match RapidMiner with Python for low latency (50-100 ms) use cases, such as making large amounts of sensor data. Also take advantage of the new feature-adapted operator to customize data with custom features by creating anomaly detection models for devices, modeling database-based physical behavior, and more.
– Security enhancements: Support for rootless Docker mode together with enhanced security in Kubernetes environments raises our overall security standards. Tank platform protection is also improved by regularly updating Docker images with the latest secure components.
– Time series prediction: Automate the prediction of future values for univariate time series based on historical data in RapidMiner Go. Follow advanced and seasonal trends when it comes to predicting sales or staffing needs and use intuitive visualizations to compare results from competing models.
NLP Extension: Take advantage of the new RapidMiner natural language extension to extract tags from a voice tag and identify people, cities, organizations and other entities in free text. This is usually used as a preprocessing method to determine the content of documents, website text, etc.
Error correction
– The main Pivot operator now works as expected within the SparkRM operator.
– Updated heuristics to read the Hive table in Radoop Sparkjobb to avoid crashing Spark jobs when there are hidden Hive test directories.