Snowball

• Snowball: Snowball is an Eager Model-based Semi-Supervised Learning-based Relation Recognition Algorithm designed for Information Extraction Tasks that involve the extraction of Binary Relations.
  • Context:
    • Automated Text Annotation includes Named Entity Recognition.
    • Uses a [[Five-Tuple_Lexically-based_Relation_Recognition_Classifier?]] Classification Model representation.
    • Uses Clustering to generalize Patterns.
    • Uses Bootstrapping to benefit from Unlabeled Data.
    • Uses Pattern Precision and a minimum precision [[Threshold?]] to stop [[Pattern_Generation?]].
    • Optimized for One-to-One Relations and One-to-Many Relations.
  • Examples:
    • http://snowball.cs.columbia.edu/
  • See: PPLRE Snowball, Snowball Internal Parameters, Snowball Algorithm Description

---

---

References

• [Agichtein and Gravano, 2000] => E. Agichtein and L. Gravano. (2000). Snowball: Extracting Relations from Large Plain-Text Collections. In Proc. of the 5th ACM Int. Conf. on Digital Libraries (DL-2000). (tech report.pdf)
• [Yu and Agichtein, 2003] => H. Yu and E. Agichtein. (2003). Extracting Synonymous Gene and Protein Terms from Biological Literature. In Proc. of the 11th Int. Conf. on Intelligent Systems for Molecular Biology (ISMB-2003). (paper.pdf)
• [Xia, 2006] => L. Xia. (2006). http://www.dcs.shef.ac.uk/intranet/teaching/projects/archive/msc2006/abs/acp05lx.htm">Adaptive Relationship Extraction by Machine Learning. Masters Thesis, Sheffield University.