Training Instance Selection Task

A Training Instance Selection Task is a biased sampling task of training instances from a training set.

• Context:
  • It can be solved by a Training Instance Selection System (that implements a Training Instance Selection Algorithm).
• See: Uncertainty Sampling.

---

---

References

2013

• (Liu & Motoda, 2013) ⇒ Huan Liu, and Hiroshi Motoda, eds. (2013). “Instance Selection and Construction for Data Mining." Vol. 608 . Springer Science & Business Media,

2006

• (Chen et al., 2006) ⇒ Yixin Chen, Jinbo Bi, and James Z. Wang. (2006). “MILES: Multiple-instance Learning via Embedded Instance Selection." Pattern Analysis and Machine Intelligence, IEEE Transactions on 28, no. 12

2002

• (Brighton & Mellish, 2002) ⇒ Henry Brighton, and Chris Mellish. (2002). “Advances in Instance Selection for Instance-based Learning Algorithms." Data mining and knowledge discovery 6, no. 2

---