Balancing Predictive Power with Business Requirements and Explainability in Automated Decisions

SUNNYVALE, Calif. - Feb. 29, 2024 - PRLog -- Sparkling Logic today announced the launch of Interactive Tree Model, a new component of the SMARTS™ decision management platform that allows users to interact with data to build, test, and deploy decision trees.

A decision tree is a type of machine learning which maps out possible consequences in a tree-like fashion with nodes and branches. In automated decisions, decision trees are especially beneficial for organizations that want to leverage machine learning in the context of business and regulatory constraints. Use cases include prescreening, fraud detection, and credit scoring in financial services, underwriting and claims processing in insurance, and marketing channel optimization across industries.

Embedded in the SMARTS™ decision management platform, Interactive Tree Model allows decision trees to be added into any automated decision project and follow the same testing, deployment, and management cycle. Once added, users choose the algorithm type and build out the nodes in an interactive manner. At each node, Interactive Tree Model will automatically provide branching recommendations which can be further customized according to the user's goals.

To improve predictive power, Interactive Tree Model can ingest large amounts of training data and handle complexity such as having multiple target variables. In addition, decision trees can be used in conjunction with any decision logic representation that SMARTS™ supports, including lookup models, scorecards, and other predictive models. As a result, organizations can balance predictability, explainability, and compliance in a completely transparent manner.

"Deployment and managing business constraints are the most common issues preventing businesses that we come across from leveraging machine learning effectively," said Jihae Hwang, VP of Data Science at Sparkling Logic. "Interactive Tree Model provides a simple path for business analysts to not only simulate, experiment, and fully deploy decision trees into their automated decisions but also explicitly refine machine learning results according to evolving business requirements."

Sparkling Logic will host a live demo on March 21. Learn more and register here.

About Sparkling Logic

Sparkling Logic Inc. is a Silicon Valley-based company dedicated to helping organizations automate and optimize key decisions in daily business operations and customer interactions in a low-code, no-code environment. Our customers include global leaders in financial services, insurance, healthcare, retail, utilities, and IoT.

SMARTS™ by Sparkling Logic is a modern, enterprise-level, application-agnostic, decision-management platform that enables non-technical analysts and SMEs to easily manage business-critical decisions, whether explicit or AI-driven, with minimal IT resources.

Learn more at www.sparklinglogic.com.

Contact
Sparkling Logic
***@sparklinglogic.com
Photos: (Click photo to enlarge)




Source: Sparkling Logic, Inc.

Read Full Story - Sparkling Logic Launches Interactive Decision Tree Modeling in SMARTS™ | More news from this source

Press release distribution by PRLog