Edit Distance

From SONIVIS:Wiki

Edit Distance This metric is not implemented in any version. You can report this as a Bug. This metric belongs to a content element. This metric is based on articles. For more information see configuration.

Edit distance is the amount of changes like the deletion of words, their movement or addition etc. between revisions. As revisions belong to a page and we consider time in this calculation we can measure this metric for a page in a dynamic model.

Objective

The edit distance describes the difference between two states of a page (revisions). This measure can point out the quantity of a made change and also the activity of an author regarding her/his content contribution with this revision. Summing all edit distances made by an single author describes the measure Edit Only.

Explanation (optional)

Within a page text can be changed. To define this change between two versions of a text the edit distance can be used. With this measure we count the words added, deleted, replaced, displaced etc. between two versions of text. It is a characteristics of a page in a dynamic model, to see how a page changes.

Like the authors B. T. Adler, L. de Alfaro, I. Pye and V. Raman^[1] say these edit distances can be used to find the “direction” of a change. This is used to calculate the Edit size, Edit longevity and the Edit Only.

Calculation

B. T. Adler, L. de Alfaro, I. Pye and V. Raman^[1] calculate the edit distance as following:

(a detailed treatment in "A content-driven reputation system for the Wikipedia"^[1])

d(v_i,v_j) for , is the edit distance between v_i and v_j, and measures how much change (word additions, deletions, replacements, displacements etc.) there has been in going from v_i to v_j. Defining d(r_i) = d(v_{i − 1},vi) for the edit contribution made in a revision r_i.

I(v_i,v_j) is the number of words that are inserted

D(v_i,v_j) is the number of words that are deleted

M(v_i,v_j) is the number of words that are moved

The edit distance can be computed using the block analysis.

Version	Edit distance
9 This is a text.
10 This is a lovely text.	d(9,10)= 1 word
11 This is a very nice and lovely text.	d(10,11)= 3 words

d(9,11)= 1 word + 3 words = 4 words

Note: A different calculation is also possible: sum the changed characters to define a edit distance.

Reference

• Presentation at WikiSym '08

  @inproceedings{Adler2008,
       title = {Measuring Author Contribution to the Wikipedia},
	address = {Porto, Portugal},
	author = {B. Thomas Adler and Luca de Alfaro and Ian Pye and Vishwanath Raman},
	booktitle = {WikiSym ’08},
	month = {September},
	url = {http://www.wikisym.org/ws2008/proceedings/research%20papers/18500027.pdf},
	year = {2008},
	keywords = {Wikipedia contribution edit_longevity quality wiki}
  }

  @inproceedings{citeulike:1291537,
       address = {New York, NY, USA},
       author = {Adler, Thomas  B.  and de Alfaro, Luca  },
       booktitle = {WWW '07: Proceedings of the 16th international conference on World Wide Web},
       doi = {http://dx.doi.org/10.1145/1242572.1242608},
       keywords = {content-driven, reputation, wikipedia},
       pages = {261--270},
       publisher = {ACM Press},
       title = {A content-driven reputation system for the wikipedia},
       year = {2007}
  }