Long-Distance Relationship

A long-distance relationship is a relationship between two or more items that are far from other according to some distance measure.

AKA: Long-Range Dependency.
Context:
- It can be represented by a Long-Range Predictor Feature.
Example(s):
- a Long-Range Word Mentions Relationship.
- The (sole) mentioning of an organism (e.g. E.coli) in the first sentence of a document can influence the disambiguation of the meaning of a protein mentions in subsequent sentences of the document.
- …
Counter-Example(s):
- a Short-Distance Relationship/Short Range Dependency.
See: Global Relationship, Global Feature, Context Predictor Feature.

References

(Zelenko et al., 2003) ⇒ Dmitry Zelenko, Chinatsu Aone, and Anthony Richardella. (2003). “Kernel Methods for Relation Extraction.” In: Journal of Machine Learning Research, 3.
- QUOTE: HMMs are mostly appropriate for modeling local and flat problems. Relation extraction often involves modeling long range dependencies, for which HMM methodology is not directly applicable. Several probabilistic frameworks for modeling sequential data have recently been introduced to alleviate for HMM restrictions.