Wrapper Induction Algorithm

A Wrapper Induction Algorithm is a relation mention recognition algorithm that can be implemented by a wrapper induction system (to solve a wrapper induction task to induce wrapper patterns).

• AKA: Extraction Rule Learner.
• Context:
  • It can (typically) be applied to semi-structured artifacts such as HTML pages generated by CGI scripts.
  • It can generalize over a set of examples.
• See: Wrapper Pattern.

---

---

References

• http://publications.csail.mit.edu/abstracts/abstracts07/gabi-wrapster/gabi-wrapster.html
  • Issues of wrapper generation systems include scalability to many sites and flexibility to hangle site changes.
  • Scalability concerns the variations of layout and format between many sites.
  • Flexibility concerns the robustness of the wrappers to frequent layout and format change.

2007

• Gabriel Zaccak. (2007). “Wrapster : semi-automatic wrapper generation for semi-structured websites." Thesis (S.M.)--Massachusetts Institute of Technology.

1999

• S. Soderland. (1999). “Learning Information Extraction Rules for Semistructured and Free Text.” In: Machine Learning 1999.
  • Wrapper Induction Algorithm, Relation Mention Pattern.

1998

• Dayne Freitag. (1998). “Information Extraction from HTML: Application of a General Machine Learning Approach.” In: Proceedings of AAAI/IAAI 1998.
  • Wrapper Induction Algorithm, Relation Mention Pattern.

1997

• Nicholas Kushmerick, D. S. Weld, and R. B. Doorenhos. (1997). “Wrapper Induction for Information Extraction.” In: International Joint Conference on Artificial Intelligence (IJCAI 1997).
  • Wrapper Induction Algorithm, Relation Mention Pattern.
• Nicholas Kushmerick. (1997). “Wrapper induction for Information Extraction." PhD Thesis, University of Washington.
  • A wrapper is a procedure for extracting a particular resource's content. Unfortunately

hand-coding wrappers is tedious We introduce wrapper induction
a technique for automatically constructing wrappers.

• • Our goal is to automatically construct wrappers Since a wrapper is simply a computer program

we are essentially trying to do automatic programming. Of course
in general automatic programming is very di cult. So
as suggested earlier
we follow standard practice and proceed by isolating particular classes of programs for which effective automatic techniques can be developed.

• • Finally

a wrapper is a procedure for extracting information from a particular resource. Formally
a wrapper takes as input a query response
and returns as output the set of tuples describing the response s information content.

• • Finally

a wrapper is a procedure for extracting the relational content from a page while discarding the irrelevant text.

• • Figure 2.1 A fictitious Internet site providing information about countries and their telephone country codes

      
       Congo 242

       Egypt 20

       Belize 501

       Spain 34

     

---