IBM Machine Learning for z/OS Announced
IBM Machine Learning for z/OS, transforms the platform into a cognitive learning system through continuous feedback, which simplifies model retraining.
Organizations wanting to use machine learning to help solve their toughest business questions are challenged by the complexity of the tools involved and the skills needed. Business rules are changing, application developers can’t always keep up and models constantly need to be tuned. To stay ahead of this rapid pace, IBM today announces machine learning capabilities for IBM z Systems.
IBM Machine Learning for z/OS transforms the platform into a cognitive learning system through continuous feedback, which simplifies model retraining. Models can improve as they are exposed to more data and human intelligence is augmented to help organizations optimize recommendations and decisions.
Traditional machine learning requires significant development, deployment, management and human intervention. The IBM approach focuses on quick model development, continuous auditing and proactive notification, and easy management. The information is processed securely and in place. This gives clients the ability to drive better business results, and to identify and minimize risk.
Many industries on z Systems can benefit from machine learning, including finance and banking, healthcare, and airlines and transportation. Benefits and business values include improved revenue generation, minimizing cost and risk through decision impact, business evolution by following market changes and increased productivity with self-learning models that become more efficient. This is a cost-effective, low-latency and high-security environment.
IBM Machine Learning for z/OS allows clients to gain advantage from z Systems infrastructure, people and processes; leverage data in place while combining structured and unstructured data from various data sources and access live transactional data.
Cognitive and machine learning capabilities work together to accelerate insight creation by making machine learning simpler and more reliable to develop, deploy, and maintain with sustainable and accurate results. It does so through:
- Integrating tools and functions needed for machine learning
- Automating management of the end-to-end machine learning workflow
- Providing a platform for better collaboration across roles, including data scientist, data engineer, business analyst and application developers
- Infusing cognitive capabilities into the workflow to help determine when model results need to be tuned and suggest updates or changes
A simple framework manages the entire machine learning workflow. Key functions are delivered through GUI, RESful API and programming APIs. Technology choices available include Scala language and SparkML framework.
Together with Hybrid Transactional/Analytical Processing and Apache Spark, clients are able to get insights on the world’s operational data such as advanced fraud detection. IBM Machine Learning for z/OS has easy model deployment user interface and API to deploy models into the Spark engine.
Existing machine learning models can be imported and/or deployed to IBM Machine Learning for z/OS, allowing you to manage and monitor them in the framework. New models can be created in the integrated Jupyter notebook environment infused with IBM cognitive capabilities to help intelligently select the best model from many possible candidates. Models also can be created less programmatically via a few guided steps through a visual model builder. This model will be ready for continuous training, including a feedback loop that allows the client to ingest new data and generate new predictions for further improvements. IBM will work to identify specific areas for the client where value can be derived from machine learning.
To get started, clients must be running IBM zEnterprise EC12 or above on z Systems servers or Linux. For z/OS clients, V2.1 or above is needed. Learn more about IBM Machine Learning for z/OS and get started today.
Valerie Dennis Craven is a Minneapolis-based writer and editor.
See more by Valerie Dennis Craven