Context-Dependent Similarity
This work addresses the problem of defining similarity measures in machine learning, but it appears incremental as it builds on existing mechanisms without broad SOTA impact.
The paper studied and compared attribute weighting and differential weighting for context-dependent similarity measures, proposing a new dissimilarity measure based on subset size and showing that attribute weighting measures are metrics while differential weighting ones are usually non-metric.
Attribute weighting and differential weighting, two major mechanisms for computing context-dependent similarity or dissimilarity measures are studied and compared. A dissimilarity measure based on subset size in the context is proposed and its metrization and application are given. It is also shown that while all attribute weighting dissimilarity measures are metrics differential weighting dissimilarity measures are usually non-metric.