Document (#32579)

Author

Chen, Z.
Fu, B.

Title

On the complexity of Rocchio's similarity-based relevance feedback algorithm

Source

Journal of the American Society for Information Science and Technology. 58(2007) no.10, S.1392-1400

Year

2007

Abstract

Rocchio's similarity-based relevance feedback algorithm, one of the most important query reformation methods in information retrieval, is essentially an adaptive learning algorithm from examples in searching for documents represented by a linear classifier. Despite its popularity in various applications, there is little rigorous analysis of its learning complexity in literature. In this article, the authors prove for the first time that the learning complexity of Rocchio's algorithm is O(d + d**2(log d + log n)) over the discretized vector space {0, ... , n - 1 }**d when the inner product similarity measure is used. The upper bound on the learning complexity for searching for documents represented by a monotone linear classifier (q, 0) over {0, ... , n - 1 }d can be improved to, at most, 1 + 2k (n - 1) (log d + log(n - 1)), where k is the number of nonzero components in q. Several lower bounds on the learning complexity are also obtained for Rocchio's algorithm. For example, the authors prove that Rocchio's algorithm has a lower bound Omega((d über 2)log n) on its learning complexity over the Boolean vector space {0,1}**d.

Theme

Retrievalalgorithmen

Object

Rocchio-Feedback-Algorithmus

Similar documents (author)

1. Chen, Y.N.; Chen, S.J.: ¬A metadata practice of the OFLA FRBR model : a case study for the National Palace Museum in Taipai (2004) 4.35

   4.3499155 = sum of:
     4.3499155 = weight(author_txt:chen in 3384) [ClassicSimilarity], result of:
       4.3499155 = fieldWeight in 3384, product of:
         1.4142135 = tf(freq=2.0), with freq of:
           2.0 = termFreq=2.0
         6.1517096 = idf(docFreq=255, maxDocs=44218)
         0.5 = fieldNorm(doc=3384)
   

2. Chen, C.C.; Chen, H.H.; Chen, K.H.: ¬The design of the XML/Metadata management system (2000) 4.00

   3.9956524 = sum of:
     3.9956524 = weight(author_txt:chen in 4633) [ClassicSimilarity], result of:
       3.9956524 = fieldWeight in 4633, product of:
         1.7320508 = tf(freq=3.0), with freq of:
           3.0 = termFreq=3.0
         6.1517096 = idf(docFreq=255, maxDocs=44218)
         0.375 = fieldNorm(doc=4633)
   

3. Chen, W.Y.: Observations on cataloguing and classification (1991) 3.84

   3.8448186 = sum of:
     3.8448186 = weight(author_txt:chen in 4184) [ClassicSimilarity], result of:
       3.8448186 = fieldWeight in 4184, product of:
         1.0 = tf(freq=1.0), with freq of:
           1.0 = termFreq=1.0
         6.1517096 = idf(docFreq=255, maxDocs=44218)
         0.625 = fieldNorm(doc=4184)
   

4. Chen, H.: Knowledge-based document retrieval : framework and design (1992) 3.84

   3.8448186 = sum of:
     3.8448186 = weight(author_txt:chen in 5283) [ClassicSimilarity], result of:
       3.8448186 = fieldWeight in 5283, product of:
         1.0 = tf(freq=1.0), with freq of:
           1.0 = termFreq=1.0
         6.1517096 = idf(docFreq=255, maxDocs=44218)
         0.625 = fieldNorm(doc=5283)
   

5. Chen, P.S.: On inference rules of logic-based information retrieval systems (1994) 3.84

   3.8448186 = sum of:
     3.8448186 = weight(author_txt:chen in 6731) [ClassicSimilarity], result of:
       3.8448186 = fieldWeight in 6731, product of:
         1.0 = tf(freq=1.0), with freq of:
           1.0 = termFreq=1.0
         6.1517096 = idf(docFreq=255, maxDocs=44218)
         0.625 = fieldNorm(doc=6731)
   

Similar documents (content)

1. Efron, M.: Query expansion and dimensionality reduction : Notions of optimality in Rocchio relevance feedback and latent semantic indexing (2008) 0.20

   0.20336226 = sum of:
     0.20336226 = product of:
       0.8473427 = sum of:
         0.016096456 = weight(abstract_txt:documents in 2020) [ClassicSimilarity], result of:
           0.016096456 = score(doc=2020,freq=1.0), product of:
             0.0499926 = queryWeight, product of:
               1.1189836 = boost
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.010840454 = queryNorm
             0.32197678 = fieldWeight in 2020, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.078125 = fieldNorm(doc=2020)
         0.047776733 = weight(abstract_txt:relevance in 2020) [ClassicSimilarity], result of:
           0.047776733 = score(doc=2020,freq=3.0), product of:
             0.07159061 = queryWeight, product of:
               1.3390565 = boost
               4.931848 = idf(docFreq=866, maxDocs=44218)
               0.010840454 = queryNorm
             0.6673603 = fieldWeight in 2020, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               4.931848 = idf(docFreq=866, maxDocs=44218)
               0.078125 = fieldNorm(doc=2020)
         0.050991446 = weight(abstract_txt:space in 2020) [ClassicSimilarity], result of:
           0.050991446 = score(doc=2020,freq=2.0), product of:
             0.08558686 = queryWeight, product of:
               1.4641124 = boost
               5.3924384 = idf(docFreq=546, maxDocs=44218)
               0.010840454 = queryNorm
             0.5957859 = fieldWeight in 2020, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               5.3924384 = idf(docFreq=546, maxDocs=44218)
               0.078125 = fieldNorm(doc=2020)
         0.083655156 = weight(abstract_txt:feedback in 2020) [ClassicSimilarity], result of:
           0.083655156 = score(doc=2020,freq=3.0), product of:
             0.10400166 = queryWeight, product of:
               1.6139535 = boost
               5.9443145 = idf(docFreq=314, maxDocs=44218)
               0.010840454 = queryNorm
             0.8043636 = fieldWeight in 2020, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               5.9443145 = idf(docFreq=314, maxDocs=44218)
               0.078125 = fieldNorm(doc=2020)
         0.090163745 = weight(abstract_txt:vector in 2020) [ClassicSimilarity], result of:
           0.090163745 = score(doc=2020,freq=2.0), product of:
             0.1251498 = queryWeight, product of:
               1.7704593 = boost
               6.5207376 = idf(docFreq=176, maxDocs=44218)
               0.010840454 = queryNorm
             0.7204466 = fieldWeight in 2020, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               6.5207376 = idf(docFreq=176, maxDocs=44218)
               0.078125 = fieldNorm(doc=2020)
         0.5586592 = weight(abstract_txt:rocchio's in 2020) [ClassicSimilarity], result of:
           0.5586592 = score(doc=2020,freq=1.0), product of:
             0.7219329 = queryWeight, product of:
               6.7234054 = boost
               9.905128 = idf(docFreq=5, maxDocs=44218)
               0.010840454 = queryNorm
             0.7738381 = fieldWeight in 2020, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.905128 = idf(docFreq=5, maxDocs=44218)
               0.078125 = fieldNorm(doc=2020)
       0.24 = coord(6/25)
   

2. Ruiz, M.E.; Srinivasan, P.: Combining machine learning and hierarchical indexing structures for text categorization (2001) 0.18

   0.17890862 = sum of:
     0.17890862 = product of:
       1.1181788 = sum of:
         0.104828805 = weight(abstract_txt:classifier in 1595) [ClassicSimilarity], result of:
           0.104828805 = score(doc=1595,freq=1.0), product of:
             0.15438975 = queryWeight, product of:
               1.9664373 = boost
               7.24254 = idf(docFreq=85, maxDocs=44218)
               0.010840454 = queryNorm
             0.6789881 = fieldWeight in 1595, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.24254 = idf(docFreq=85, maxDocs=44218)
               0.09375 = fieldNorm(doc=1595)
         0.12553489 = weight(abstract_txt:learning in 1595) [ClassicSimilarity], result of:
           0.12553489 = score(doc=1595,freq=2.0), product of:
             0.19929884 = queryWeight, product of:
               3.8697586 = boost
               4.750873 = idf(docFreq=1038, maxDocs=44218)
               0.010840454 = queryNorm
             0.6298827 = fieldWeight in 1595, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.750873 = idf(docFreq=1038, maxDocs=44218)
               0.09375 = fieldNorm(doc=1595)
         0.21742411 = weight(abstract_txt:algorithm in 1595) [ClassicSimilarity], result of:
           0.21742411 = score(doc=1595,freq=2.0), product of:
             0.287431 = queryWeight, product of:
               4.647274 = boost
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.010840454 = queryNorm
             0.7564393 = fieldWeight in 1595, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.09375 = fieldNorm(doc=1595)
         0.67039096 = weight(abstract_txt:rocchio's in 1595) [ClassicSimilarity], result of:
           0.67039096 = score(doc=1595,freq=1.0), product of:
             0.7219329 = queryWeight, product of:
               6.7234054 = boost
               9.905128 = idf(docFreq=5, maxDocs=44218)
               0.010840454 = queryNorm
             0.9286057 = fieldWeight in 1595, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.905128 = idf(docFreq=5, maxDocs=44218)
               0.09375 = fieldNorm(doc=1595)
       0.16 = coord(4/25)
   

3. Mengle, S.S.R.; Goharian, N.: Ambiguity measure feature-selection algorithm (2009) 0.15

   0.14690474 = sum of:
     0.14690474 = product of:
       0.52465975 = sum of:
         0.011283036 = weight(abstract_txt:most in 2804) [ClassicSimilarity], result of:
           0.011283036 = score(doc=2804,freq=1.0), product of:
             0.04577653 = queryWeight, product of:
               1.0707604 = boost
               3.943693 = idf(docFreq=2328, maxDocs=44218)
               0.010840454 = queryNorm
             0.24648081 = fieldWeight in 2804, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.943693 = idf(docFreq=2328, maxDocs=44218)
               0.0625 = fieldNorm(doc=2804)
         0.018211061 = weight(abstract_txt:documents in 2804) [ClassicSimilarity], result of:
           0.018211061 = score(doc=2804,freq=2.0), product of:
             0.0499926 = queryWeight, product of:
               1.1189836 = boost
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.010840454 = queryNorm
             0.36427513 = fieldWeight in 2804, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.0625 = fieldNorm(doc=2804)
         0.028845116 = weight(abstract_txt:space in 2804) [ClassicSimilarity], result of:
           0.028845116 = score(doc=2804,freq=1.0), product of:
             0.08558686 = queryWeight, product of:
               1.4641124 = boost
               5.3924384 = idf(docFreq=546, maxDocs=44218)
               0.010840454 = queryNorm
             0.3370274 = fieldWeight in 2804, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.3924384 = idf(docFreq=546, maxDocs=44218)
               0.0625 = fieldNorm(doc=2804)
         0.051004313 = weight(abstract_txt:vector in 2804) [ClassicSimilarity], result of:
           0.051004313 = score(doc=2804,freq=1.0), product of:
             0.1251498 = queryWeight, product of:
               1.7704593 = boost
               6.5207376 = idf(docFreq=176, maxDocs=44218)
               0.010840454 = queryNorm
             0.4075461 = fieldWeight in 2804, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               6.5207376 = idf(docFreq=176, maxDocs=44218)
               0.0625 = fieldNorm(doc=2804)
         0.15626955 = weight(abstract_txt:classifier in 2804) [ClassicSimilarity], result of:
           0.15626955 = score(doc=2804,freq=5.0), product of:
             0.15438975 = queryWeight, product of:
               1.9664373 = boost
               7.24254 = idf(docFreq=85, maxDocs=44218)
               0.010840454 = queryNorm
             1.0121757 = fieldWeight in 2804, product of:
               2.236068 = tf(freq=5.0), with freq of:
                 5.0 = termFreq=5.0
               7.24254 = idf(docFreq=85, maxDocs=44218)
               0.0625 = fieldNorm(doc=2804)
         0.14494942 = weight(abstract_txt:algorithm in 2804) [ClassicSimilarity], result of:
           0.14494942 = score(doc=2804,freq=2.0), product of:
             0.287431 = queryWeight, product of:
               4.647274 = boost
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.010840454 = queryNorm
             0.5042929 = fieldWeight in 2804, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.0625 = fieldNorm(doc=2804)
         0.11409727 = weight(abstract_txt:complexity in 2804) [ClassicSimilarity], result of:
           0.11409727 = score(doc=2804,freq=1.0), product of:
             0.3087328 = queryWeight, product of:
               4.816404 = boost
               5.913062 = idf(docFreq=324, maxDocs=44218)
               0.010840454 = queryNorm
             0.36956638 = fieldWeight in 2804, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.913062 = idf(docFreq=324, maxDocs=44218)
               0.0625 = fieldNorm(doc=2804)
       0.28 = coord(7/25)
   

4. Borodin, Y.; Polishchuk, V.; Mahmud, J.; Ramakrishnan, I.V.; Stent, A.: Live and learn from mistakes : a lightweight system for document classification (2013) 0.14

   0.14410289 = sum of:
     0.14410289 = product of:
       0.51465315 = sum of:
         0.022303902 = weight(abstract_txt:documents in 2722) [ClassicSimilarity], result of:
           0.022303902 = score(doc=2722,freq=3.0), product of:
             0.0499926 = queryWeight, product of:
               1.1189836 = boost
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.010840454 = queryNorm
             0.44614407 = fieldWeight in 2722, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.0625 = fieldNorm(doc=2722)
         0.028845116 = weight(abstract_txt:space in 2722) [ClassicSimilarity], result of:
           0.028845116 = score(doc=2722,freq=1.0), product of:
             0.08558686 = queryWeight, product of:
               1.4641124 = boost
               5.3924384 = idf(docFreq=546, maxDocs=44218)
               0.010840454 = queryNorm
             0.3370274 = fieldWeight in 2722, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.3924384 = idf(docFreq=546, maxDocs=44218)
               0.0625 = fieldNorm(doc=2722)
         0.038638663 = weight(abstract_txt:feedback in 2722) [ClassicSimilarity], result of:
           0.038638663 = score(doc=2722,freq=1.0), product of:
             0.10400166 = queryWeight, product of:
               1.6139535 = boost
               5.9443145 = idf(docFreq=314, maxDocs=44218)
               0.010840454 = queryNorm
             0.37151965 = fieldWeight in 2722, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.9443145 = idf(docFreq=314, maxDocs=44218)
               0.0625 = fieldNorm(doc=2722)
         0.036546316 = weight(abstract_txt:over in 2722) [ClassicSimilarity], result of:
           0.036546316 = score(doc=2722,freq=3.0), product of:
             0.079538606 = queryWeight, product of:
               1.7286437 = boost
               4.244485 = idf(docFreq=1723, maxDocs=44218)
               0.010840454 = queryNorm
             0.45947897 = fieldWeight in 2722, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               4.244485 = idf(docFreq=1723, maxDocs=44218)
               0.0625 = fieldNorm(doc=2722)
         0.051004313 = weight(abstract_txt:vector in 2722) [ClassicSimilarity], result of:
           0.051004313 = score(doc=2722,freq=1.0), product of:
             0.1251498 = queryWeight, product of:
               1.7704593 = boost
               6.5207376 = idf(docFreq=176, maxDocs=44218)
               0.010840454 = queryNorm
             0.4075461 = fieldWeight in 2722, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               6.5207376 = idf(docFreq=176, maxDocs=44218)
               0.0625 = fieldNorm(doc=2722)
         0.13232541 = weight(abstract_txt:learning in 2722) [ClassicSimilarity], result of:
           0.13232541 = score(doc=2722,freq=5.0), product of:
             0.19929884 = queryWeight, product of:
               3.8697586 = boost
               4.750873 = idf(docFreq=1038, maxDocs=44218)
               0.010840454 = queryNorm
             0.66395473 = fieldWeight in 2722, product of:
               2.236068 = tf(freq=5.0), with freq of:
                 5.0 = termFreq=5.0
               4.750873 = idf(docFreq=1038, maxDocs=44218)
               0.0625 = fieldNorm(doc=2722)
         0.20498943 = weight(abstract_txt:algorithm in 2722) [ClassicSimilarity], result of:
           0.20498943 = score(doc=2722,freq=4.0), product of:
             0.287431 = queryWeight, product of:
               4.647274 = boost
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.010840454 = queryNorm
             0.71317786 = fieldWeight in 2722, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.0625 = fieldNorm(doc=2722)
       0.28 = coord(7/25)
   

5. Widyantoro, D.H.; Ioerger, T.R.; Yen, J.: Learning user Interest dynamics with a three-descriptor representation (2001) 0.14

   0.14229059 = sum of:
     0.14229059 = product of:
       0.71145296 = sum of:
         0.018211061 = weight(abstract_txt:documents in 5185) [ClassicSimilarity], result of:
           0.018211061 = score(doc=5185,freq=2.0), product of:
             0.0499926 = queryWeight, product of:
               1.1189836 = boost
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.010840454 = queryNorm
             0.36427513 = fieldWeight in 5185, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.1213026 = idf(docFreq=1949, maxDocs=44218)
               0.0625 = fieldNorm(doc=5185)
         0.066924125 = weight(abstract_txt:feedback in 5185) [ClassicSimilarity], result of:
           0.066924125 = score(doc=5185,freq=3.0), product of:
             0.10400166 = queryWeight, product of:
               1.6139535 = boost
               5.9443145 = idf(docFreq=314, maxDocs=44218)
               0.010840454 = queryNorm
             0.6434909 = fieldWeight in 5185, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               5.9443145 = idf(docFreq=314, maxDocs=44218)
               0.0625 = fieldNorm(doc=5185)
         0.076895684 = weight(abstract_txt:similarity in 5185) [ClassicSimilarity], result of:
           0.076895684 = score(doc=5185,freq=2.0), product of:
             0.1495021 = queryWeight, product of:
               2.3699553 = boost
               5.8191514 = idf(docFreq=356, maxDocs=44218)
               0.010840454 = queryNorm
             0.51434517 = fieldWeight in 5185, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               5.8191514 = idf(docFreq=356, maxDocs=44218)
               0.0625 = fieldNorm(doc=5185)
         0.10249472 = weight(abstract_txt:algorithm in 5185) [ClassicSimilarity], result of:
           0.10249472 = score(doc=5185,freq=1.0), product of:
             0.287431 = queryWeight, product of:
               4.647274 = boost
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.010840454 = queryNorm
             0.35658893 = fieldWeight in 5185, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.0625 = fieldNorm(doc=5185)
         0.44692734 = weight(abstract_txt:rocchio's in 5185) [ClassicSimilarity], result of:
           0.44692734 = score(doc=5185,freq=1.0), product of:
             0.7219329 = queryWeight, product of:
               6.7234054 = boost
               9.905128 = idf(docFreq=5, maxDocs=44218)
               0.010840454 = queryNorm
             0.6190705 = fieldWeight in 5185, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.905128 = idf(docFreq=5, maxDocs=44218)
               0.0625 = fieldNorm(doc=5185)
       0.2 = coord(5/25)