Docstoc

6.9 Meta-Search and Combining Multiple Rankings

Document Sample
6.9 Meta-Search and Combining Multiple Rankings Powered By Docstoc
					6.9 Meta-Search and Combining
       Multiple Rankings

                   Date:2009/11/30
            Speaker:Hsin Lan, Wang
                   Outline

•   Meta-Search Engine
•   Key Operation in Meta-Search
•   Combination Using Similarity Scores
•   Combination Using Rank Positions
           Meta-Search Engine

• Several search engines can be used
  together to produce a meta-search engine.
• It is a search system that does not have its
  own database of Web pages.
• It answers the user query by combining
  the results of some other search engines.
          Meta-Search Engine



• Receiving a query from the user through
  the search interface.
• Meta-search engine submits the query to
  the underlying search engines.
• The returned results from all these search
  engines are then combined and sent to the
  user.
           Meta-Search Engine

• Intuitive appeals:
  Increases the search coverage of the Web
    • Each individual search engine may only cover a
      small portion of the Web.
    • Using only one search engine will never see those
      relevant pages that are not covered by the search
      engine.
  Improve the search effectiveness
    • By combining the results from multiple search
      engines, their biases can be reduced.
    • The search precision can be improved.
     Key Operation in Meta-Search

• The key operation in meta-search is to
  combine the ranked results from the
  component search engines to produce a
  single ranking.
     Key Operation in Meta-Search

• First task:
  To identify whether two pages from different
   search engines are the same, which facilitates
   combination and duplicate removal.
• Second task:
  To combine the ranked results from individual
   search engines to produce a single ranking.
  Two main classes:
    • Combination Using Similarity Scores
    • Combination Using Rank Positions
 Combination Using Similarity Scores

• D={d1, d2, …, dN}:The set of candidate
  documents to be ranked.
• There are k underlying systems.
• The ranking from system or technique i
  gives document dj the similarity score, sij.
Combination Using Similarity Scores
Combination Using Similarity Scores
  Combination Using Rank Positions

• Borda ranking:
  The top candidate receives n points, the
   second candidate receives n-1 points, and so
   on.
  The points from all voters are summed up to
   give the final points for each candidate.
  If there are candidates left unranked by a
   voter, the remaining points are divided evenly
   among the unranked candidates.
  Combination Using Rank Positions

• Borda ranking:



               剩下一分,分給一個沒有ranked到的文章
             剩下二加一分,分給兩個沒有ranked到的文章
               1/1 = 1
             (2+1)/2 = 1.5
  Combination Using Rank Positions

• Condorcet ranking:
  The winner of the election is the candidate(s)
   that beats each of the other candidates in a
   pair-wise comparison.
  If a candidate is not ranked by a voter, the
   candidate loses to all other ranked candidates.
  All unranked candidates tie with one another.
  Combination Using Rank Positions
                WIN : 10
                       3
                       2
                LOSE : 1 0
• Condorcet ranking:  0
                TIE : 1
           a和b都沒有在list中,所以平手
  Combination Using Rank Positions

• Reciprocal ranking:
  Sums one over the rank of each candidate
   across all voters.
  The top candidate has the score of 1, the
   second ranked candidate has the score of ½,
   and so on.
  If a candidate is not ranked by a voter, it is
   skipped in the computation for this voter.
  Combination Using Rank Positions

• Reciprocal ranking:

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:4
posted:8/30/2012
language:Unknown
pages:16