Long-Range Predictor Feature

A Long-Range Predictor Feature is a Predictor Feature that makes use of Information that is Near to an Item represented by a Learning Record according to some Distance Function.

• AKA: Long-Range Feature.
• Context:
  • It can represent a Long-Range Relationship.
• Example(s):
  • a Predictor Feature based on information from outside the Text Window that contains the Item of the Learning Record.
  • …
• Counter-Example(s):
  • a Local Predictor Feature.
  • a Global Predictor Feature.
• See: Context Feature.

---

---

References

2003

• (Zelenko et al., 2003) ⇒ Dmitry Zelenko, Chinatsu Aone, and Anthony Richardella. (2003). “Kernel Methods for Relation Extraction.” In: Journal of Machine Learning Research, 3.
  • … We therefore are able to take advantage of long-range features in strings without enumerating the features explicitly. ...

---