18327763. REASONING WITH CONDITIONAL INDEPENDENCE GRAPHS simplified abstract (Microsoft Technology Licensing, LLC)

REASONING WITH CONDITIONAL INDEPENDENCE GRAPHS

Organization Name

Microsoft Technology Licensing, LLC

Inventor(s)

Urszula Stefania Chajewska of Camano Island WA (US)

Harsh Shrivastava of Redmond WA (US)

REASONING WITH CONDITIONAL INDEPENDENCE GRAPHS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18327763 titled 'REASONING WITH CONDITIONAL INDEPENDENCE GRAPHS

The present disclosure involves the propagation of knowledge between nodes of a feature graph to predict attribute values for various features, enabling analysis beyond direct dependencies in complex domain spaces.

• Generating a transition matrix based on correlations within the feature graph to distribute weights to an attribute matrix with known and unknown attribute values.
• Provides a computationally inexpensive approach to evaluating graphs of complex domains by considering indirectly correlated features.
• Enhances the inference or prediction of attribute values within a feature graph.

Potential Applications: This technology can be applied in various fields such as machine learning, data analysis, predictive modeling, and network analysis.

Problems Solved: This technology addresses the challenge of inferring attribute values for features in complex domain spaces with indirect correlations.

Benefits: - Improved accuracy in predicting attribute values - Enhanced analysis of feature graphs in complex domains - Cost-effective approach to evaluating complex networks

Commercial Applications: This technology can be utilized in industries such as finance, healthcare, marketing, and telecommunications for data analysis, predictive modeling, and network optimization.

Questions about the technology: 1. How does this technology improve the analysis of feature graphs in complex domains? 2. What are the potential applications of this technology in machine learning and data analysis?

Frequently Updated Research: Stay updated on the latest advancements in machine learning, data analysis, and network analysis to enhance the application of this technology.

Original Abstract Submitted

The present disclosure relates to propagating knowledge between nodes of a feature graph in inferring or otherwise predicting attribute values for various features represented within the feature graph. This enables analysis of the feature graph beyond direct dependencies and for domain spaces that are increasingly complex. The present disclosure includes generating a transition matrix based on correlations within the feature graph to determine distribution of weights to apply to an attribute matrix including a combination of known and unknown attribute values. Features described herein provide a computationally inexpensive and flexible approach to evaluating graphs of complex domains while considering combinations of features that are not necessarily directly correlated to other features.