Document (#37247)

Author

Levin, M.
Krawczyk, S.
Bethard, S.
Jurafsky, D.

Title

Citation-based bootstrapping for large-scale author disambiguation

Source

Journal of the American Society for Information Science and Technology. 63(2012) no.5, S.1030-1047

Year

2012

Abstract

We present a new, two-stage, self-supervised algorithm for author disambiguation in large bibliographic databases. In the first "bootstrap" stage, a collection of high-precision features is used to bootstrap a training set with positive and negative examples of coreferring authors. A supervised feature-based classifier is then trained on the bootstrap clusters and used to cluster the authors in a larger unlabeled dataset. Our self-supervised approach shares the advantages of unsupervised approaches (no need for expensive hand labels) as well as supervised approaches (a rich set of features that can be discriminatively trained). The algorithm disambiguates 54,000,000 author instances in Thomson Reuters' Web of Knowledge with B3 F1 of.807. We analyze parameters and features, particularly those from citation networks, which have not been deeply investigated in author disambiguation. The most important citation feature is self-citation, which can be approximated without expensive extraction of the full network. For the supervised stage, the minor improvement due to other citation features (increasing F1 from.748 to.767) suggests they may not be worth the trouble of extracting from databases that don't already have them. A lean feature set without expensive abstract and title features performs 130 times faster with about equal F1.

Theme

Informetrie
Computerlinguistik

Similar documents (content)

1. Ferreira, A.A.; Veloso, A.; Gonçalves, M.A.; Laender, A.H.F.: Self-training author name disambiguation for information scarce scenarios (2014) 0.30

   0.30153102 = sum of:
     0.30153102 = product of:
       0.94228446 = sum of:
         0.006622435 = weight(abstract_txt:from in 1292) [ClassicSimilarity], result of:
           0.006622435 = score(doc=1292,freq=1.0), product of:
             0.038336996 = queryWeight, product of:
               1.0288296 = boost
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.013482022 = queryNorm
             0.17274266 = fieldWeight in 1292, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
         0.021004213 = weight(abstract_txt:authors in 1292) [ClassicSimilarity], result of:
           0.021004213 = score(doc=1292,freq=1.0), product of:
             0.07229587 = queryWeight, product of:
               1.1535741 = boost
               4.648501 = idf(docFreq=1150, maxDocs=44218)
               0.013482022 = queryNorm
             0.2905313 = fieldWeight in 1292, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.648501 = idf(docFreq=1150, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
         0.09693533 = weight(abstract_txt:bootstrapping in 1292) [ClassicSimilarity], result of:
           0.09693533 = score(doc=1292,freq=1.0), product of:
             0.15905742 = queryWeight, product of:
               1.2099036 = boost
               9.7509775 = idf(docFreq=6, maxDocs=44218)
               0.013482022 = queryNorm
             0.6094361 = fieldWeight in 1292, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.7509775 = idf(docFreq=6, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
         0.0934445 = weight(abstract_txt:self in 1292) [ClassicSimilarity], result of:
           0.0934445 = score(doc=1292,freq=3.0), product of:
             0.15521549 = queryWeight, product of:
               2.0701506 = boost
               5.561322 = idf(docFreq=461, maxDocs=44218)
               0.013482022 = queryNorm
             0.60203075 = fieldWeight in 1292, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               5.561322 = idf(docFreq=461, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
         0.11534916 = weight(abstract_txt:author in 1292) [ClassicSimilarity], result of:
           0.11534916 = score(doc=1292,freq=5.0), product of:
             0.165808 = queryWeight, product of:
               2.4706233 = boost
               4.9778743 = idf(docFreq=827, maxDocs=44218)
               0.013482022 = queryNorm
             0.69567907 = fieldWeight in 1292, product of:
               2.236068 = tf(freq=5.0), with freq of:
                 5.0 = termFreq=5.0
               4.9778743 = idf(docFreq=827, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
         0.21485199 = weight(abstract_txt:disambiguation in 1292) [ClassicSimilarity], result of:
           0.21485199 = score(doc=1292,freq=3.0), product of:
             0.27039093 = queryWeight, product of:
               2.7323132 = boost
               7.3401785 = idf(docFreq=77, maxDocs=44218)
               0.013482022 = queryNorm
             0.7945976 = fieldWeight in 1292, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               7.3401785 = idf(docFreq=77, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
         0.08680359 = weight(abstract_txt:citation in 1292) [ClassicSimilarity], result of:
           0.08680359 = score(doc=1292,freq=2.0), product of:
             0.20055693 = queryWeight, product of:
               3.037929 = boost
               4.896717 = idf(docFreq=897, maxDocs=44218)
               0.013482022 = queryNorm
             0.43281272 = fieldWeight in 1292, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.896717 = idf(docFreq=897, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
         0.3072732 = weight(abstract_txt:supervised in 1292) [ClassicSimilarity], result of:
           0.3072732 = score(doc=1292,freq=2.0), product of:
             0.46583167 = queryWeight, product of:
               4.629918 = boost
               7.462781 = idf(docFreq=68, maxDocs=44218)
               0.013482022 = queryNorm
             0.65962285 = fieldWeight in 1292, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               7.462781 = idf(docFreq=68, maxDocs=44218)
               0.0625 = fieldNorm(doc=1292)
       0.32 = coord(8/25)
   

2. Ko, Y.; Seo, J.: Text classification from unlabeled documents with bootstrapping and feature projection techniques (2009) 0.25

   0.24700569 = sum of:
     0.24700569 = product of:
       0.88216317 = sum of:
         0.018477548 = weight(abstract_txt:large in 2452) [ClassicSimilarity], result of:
           0.018477548 = score(doc=2452,freq=1.0), product of:
             0.06637514 = queryWeight, product of:
               1.1053288 = boost
               4.454089 = idf(docFreq=1397, maxDocs=44218)
               0.013482022 = queryNorm
             0.27838057 = fieldWeight in 2452, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.454089 = idf(docFreq=1397, maxDocs=44218)
               0.0625 = fieldNorm(doc=2452)
         0.020466348 = weight(abstract_txt:approaches in 2452) [ClassicSimilarity], result of:
           0.020466348 = score(doc=2452,freq=1.0), product of:
             0.07105633 = queryWeight, product of:
               1.1436422 = boost
               4.6084785 = idf(docFreq=1197, maxDocs=44218)
               0.013482022 = queryNorm
             0.2880299 = fieldWeight in 2452, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.6084785 = idf(docFreq=1197, maxDocs=44218)
               0.0625 = fieldNorm(doc=2452)
         0.122587524 = weight(abstract_txt:unlabeled in 2452) [ClassicSimilarity], result of:
           0.122587524 = score(doc=2452,freq=2.0), product of:
             0.14763412 = queryWeight, product of:
               1.1656473 = boost
               9.394302 = idf(docFreq=9, maxDocs=44218)
               0.013482022 = queryNorm
             0.8303468 = fieldWeight in 2452, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               9.394302 = idf(docFreq=9, maxDocs=44218)
               0.0625 = fieldNorm(doc=2452)
         0.09693533 = weight(abstract_txt:bootstrapping in 2452) [ClassicSimilarity], result of:
           0.09693533 = score(doc=2452,freq=1.0), product of:
             0.15905742 = queryWeight, product of:
               1.2099036 = boost
               9.7509775 = idf(docFreq=6, maxDocs=44218)
               0.013482022 = queryNorm
             0.6094361 = fieldWeight in 2452, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.7509775 = idf(docFreq=6, maxDocs=44218)
               0.0625 = fieldNorm(doc=2452)
         0.064446315 = weight(abstract_txt:feature in 2452) [ClassicSimilarity], result of:
           0.064446315 = score(doc=2452,freq=1.0), product of:
             0.1747451 = queryWeight, product of:
               2.196529 = boost
               5.9008293 = idf(docFreq=328, maxDocs=44218)
               0.013482022 = queryNorm
             0.36880183 = fieldWeight in 2452, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.9008293 = idf(docFreq=328, maxDocs=44218)
               0.0625 = fieldNorm(doc=2452)
         0.124700174 = weight(abstract_txt:expensive in 2452) [ClassicSimilarity], result of:
           0.124700174 = score(doc=2452,freq=1.0), product of:
             0.2713424 = queryWeight, product of:
               2.7371163 = boost
               7.3530817 = idf(docFreq=76, maxDocs=44218)
               0.013482022 = queryNorm
             0.4595676 = fieldWeight in 2452, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.3530817 = idf(docFreq=76, maxDocs=44218)
               0.0625 = fieldNorm(doc=2452)
         0.43454996 = weight(abstract_txt:supervised in 2452) [ClassicSimilarity], result of:
           0.43454996 = score(doc=2452,freq=4.0), product of:
             0.46583167 = queryWeight, product of:
               4.629918 = boost
               7.462781 = idf(docFreq=68, maxDocs=44218)
               0.013482022 = queryNorm
             0.9328476 = fieldWeight in 2452, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               7.462781 = idf(docFreq=68, maxDocs=44218)
               0.0625 = fieldNorm(doc=2452)
       0.28 = coord(7/25)
   

3. Suakkaphong, N.; Zhang, Z.; Chen, H.: Disease named entity recognition using semisupervised learning and conditional random fields (2011) 0.20

   0.20278284 = sum of:
     0.20278284 = product of:
       0.6336964 = sum of:
         0.009365537 = weight(abstract_txt:from in 4367) [ClassicSimilarity], result of:
           0.009365537 = score(doc=4367,freq=2.0), product of:
             0.038336996 = queryWeight, product of:
               1.0288296 = boost
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.013482022 = queryNorm
             0.24429502 = fieldWeight in 4367, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.0625 = fieldNorm(doc=4367)
         0.017967861 = weight(abstract_txt:databases in 4367) [ClassicSimilarity], result of:
           0.017967861 = score(doc=4367,freq=1.0), product of:
             0.06514885 = queryWeight, product of:
               1.0950707 = boost
               4.4127526 = idf(docFreq=1456, maxDocs=44218)
               0.013482022 = queryNorm
             0.27579704 = fieldWeight in 4367, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.4127526 = idf(docFreq=1456, maxDocs=44218)
               0.0625 = fieldNorm(doc=4367)
         0.0261312 = weight(abstract_txt:large in 4367) [ClassicSimilarity], result of:
           0.0261312 = score(doc=4367,freq=2.0), product of:
             0.06637514 = queryWeight, product of:
               1.1053288 = boost
               4.454089 = idf(docFreq=1397, maxDocs=44218)
               0.013482022 = queryNorm
             0.39368957 = fieldWeight in 4367, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.454089 = idf(docFreq=1397, maxDocs=44218)
               0.0625 = fieldNorm(doc=4367)
         0.122587524 = weight(abstract_txt:unlabeled in 4367) [ClassicSimilarity], result of:
           0.122587524 = score(doc=4367,freq=2.0), product of:
             0.14763412 = queryWeight, product of:
               1.1656473 = boost
               9.394302 = idf(docFreq=9, maxDocs=44218)
               0.013482022 = queryNorm
             0.8303468 = fieldWeight in 4367, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               9.394302 = idf(docFreq=9, maxDocs=44218)
               0.0625 = fieldNorm(doc=4367)
         0.13708727 = weight(abstract_txt:bootstrapping in 4367) [ClassicSimilarity], result of:
           0.13708727 = score(doc=4367,freq=2.0), product of:
             0.15905742 = queryWeight, product of:
               1.2099036 = boost
               9.7509775 = idf(docFreq=6, maxDocs=44218)
               0.013482022 = queryNorm
             0.8618728 = fieldWeight in 4367, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               9.7509775 = idf(docFreq=6, maxDocs=44218)
               0.0625 = fieldNorm(doc=4367)
         0.038835704 = weight(abstract_txt:algorithm in 4367) [ClassicSimilarity], result of:
           0.038835704 = score(doc=4367,freq=1.0), product of:
             0.108908884 = queryWeight, product of:
               1.4158604 = boost
               5.705423 = idf(docFreq=399, maxDocs=44218)
               0.013482022 = queryNorm
             0.35658893 = fieldWeight in 4367, 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=4367)
         0.064446315 = weight(abstract_txt:feature in 4367) [ClassicSimilarity], result of:
           0.064446315 = score(doc=4367,freq=1.0), product of:
             0.1747451 = queryWeight, product of:
               2.196529 = boost
               5.9008293 = idf(docFreq=328, maxDocs=44218)
               0.013482022 = queryNorm
             0.36880183 = fieldWeight in 4367, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.9008293 = idf(docFreq=328, maxDocs=44218)
               0.0625 = fieldNorm(doc=4367)
         0.21727498 = weight(abstract_txt:supervised in 4367) [ClassicSimilarity], result of:
           0.21727498 = score(doc=4367,freq=1.0), product of:
             0.46583167 = queryWeight, product of:
               4.629918 = boost
               7.462781 = idf(docFreq=68, maxDocs=44218)
               0.013482022 = queryNorm
             0.4664238 = fieldWeight in 4367, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.462781 = idf(docFreq=68, maxDocs=44218)
               0.0625 = fieldNorm(doc=4367)
       0.32 = coord(8/25)
   

4. Zhao, D.; Strotmann, A.: In-text author citation analysis : feasibility, benefits, and limitations (2014) 0.19

   0.18589376 = sum of:
     0.18589376 = product of:
       0.6639063 = sum of:
         0.009365537 = weight(abstract_txt:from in 1535) [ClassicSimilarity], result of:
           0.009365537 = score(doc=1535,freq=2.0), product of:
             0.038336996 = queryWeight, product of:
               1.0288296 = boost
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.013482022 = queryNorm
             0.24429502 = fieldWeight in 1535, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.0625 = fieldNorm(doc=1535)
         0.035935722 = weight(abstract_txt:databases in 1535) [ClassicSimilarity], result of:
           0.035935722 = score(doc=1535,freq=4.0), product of:
             0.06514885 = queryWeight, product of:
               1.0950707 = boost
               4.4127526 = idf(docFreq=1456, maxDocs=44218)
               0.013482022 = queryNorm
             0.5515941 = fieldWeight in 1535, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               4.4127526 = idf(docFreq=1456, maxDocs=44218)
               0.0625 = fieldNorm(doc=1535)
         0.02970444 = weight(abstract_txt:authors in 1535) [ClassicSimilarity], result of:
           0.02970444 = score(doc=1535,freq=2.0), product of:
             0.07229587 = queryWeight, product of:
               1.1535741 = boost
               4.648501 = idf(docFreq=1150, maxDocs=44218)
               0.013482022 = queryNorm
             0.4108733 = fieldWeight in 1535, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.648501 = idf(docFreq=1150, maxDocs=44218)
               0.0625 = fieldNorm(doc=1535)
         0.029846795 = weight(abstract_txt:without in 1535) [ClassicSimilarity], result of:
           0.029846795 = score(doc=1535,freq=1.0), product of:
             0.09137786 = queryWeight, product of:
               1.2969084 = boost
               5.2260876 = idf(docFreq=645, maxDocs=44218)
               0.013482022 = queryNorm
             0.32663047 = fieldWeight in 1535, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.2260876 = idf(docFreq=645, maxDocs=44218)
               0.0625 = fieldNorm(doc=1535)
         0.14590643 = weight(abstract_txt:author in 1535) [ClassicSimilarity], result of:
           0.14590643 = score(doc=1535,freq=8.0), product of:
             0.165808 = queryWeight, product of:
               2.4706233 = boost
               4.9778743 = idf(docFreq=827, maxDocs=44218)
               0.013482022 = queryNorm
             0.87997216 = fieldWeight in 1535, product of:
               2.828427 = tf(freq=8.0), with freq of:
                 8.0 = termFreq=8.0
               4.9778743 = idf(docFreq=827, maxDocs=44218)
               0.0625 = fieldNorm(doc=1535)
         0.17542592 = weight(abstract_txt:disambiguation in 1535) [ClassicSimilarity], result of:
           0.17542592 = score(doc=1535,freq=2.0), product of:
             0.27039093 = queryWeight, product of:
               2.7323132 = boost
               7.3401785 = idf(docFreq=77, maxDocs=44218)
               0.013482022 = queryNorm
             0.64878625 = fieldWeight in 1535, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               7.3401785 = idf(docFreq=77, maxDocs=44218)
               0.0625 = fieldNorm(doc=1535)
         0.23772144 = weight(abstract_txt:citation in 1535) [ClassicSimilarity], result of:
           0.23772144 = score(doc=1535,freq=15.0), product of:
             0.20055693 = queryWeight, product of:
               3.037929 = boost
               4.896717 = idf(docFreq=897, maxDocs=44218)
               0.013482022 = queryNorm
             1.1853065 = fieldWeight in 1535, product of:
               3.8729835 = tf(freq=15.0), with freq of:
                 15.0 = termFreq=15.0
               4.896717 = idf(docFreq=897, maxDocs=44218)
               0.0625 = fieldNorm(doc=1535)
       0.28 = coord(7/25)
   

5. Strotmann, A.; Zhao, D.: Author name disambiguation : what difference does it make in author-based citation analysis? (2012) 0.13

   0.13391994 = sum of:
     0.13391994 = product of:
       0.55799973 = sum of:
         0.006622435 = weight(abstract_txt:from in 389) [ClassicSimilarity], result of:
           0.006622435 = score(doc=389,freq=1.0), product of:
             0.038336996 = queryWeight, product of:
               1.0288296 = boost
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.013482022 = queryNorm
             0.17274266 = fieldWeight in 389, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               2.7638826 = idf(docFreq=7577, maxDocs=44218)
               0.0625 = fieldNorm(doc=389)
         0.020466348 = weight(abstract_txt:approaches in 389) [ClassicSimilarity], result of:
           0.020466348 = score(doc=389,freq=1.0), product of:
             0.07105633 = queryWeight, product of:
               1.1436422 = boost
               4.6084785 = idf(docFreq=1197, maxDocs=44218)
               0.013482022 = queryNorm
             0.2880299 = fieldWeight in 389, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.6084785 = idf(docFreq=1197, maxDocs=44218)
               0.0625 = fieldNorm(doc=389)
         0.042008426 = weight(abstract_txt:authors in 389) [ClassicSimilarity], result of:
           0.042008426 = score(doc=389,freq=4.0), product of:
             0.07229587 = queryWeight, product of:
               1.1535741 = boost
               4.648501 = idf(docFreq=1150, maxDocs=44218)
               0.013482022 = queryNorm
             0.5810626 = fieldWeight in 389, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               4.648501 = idf(docFreq=1150, maxDocs=44218)
               0.0625 = fieldNorm(doc=389)
         0.16312835 = weight(abstract_txt:author in 389) [ClassicSimilarity], result of:
           0.16312835 = score(doc=389,freq=10.0), product of:
             0.165808 = queryWeight, product of:
               2.4706233 = boost
               4.9778743 = idf(docFreq=827, maxDocs=44218)
               0.013482022 = queryNorm
             0.9838388 = fieldWeight in 389, product of:
               3.1622777 = tf(freq=10.0), with freq of:
                 10.0 = termFreq=10.0
               4.9778743 = idf(docFreq=827, maxDocs=44218)
               0.0625 = fieldNorm(doc=389)
         0.17542592 = weight(abstract_txt:disambiguation in 389) [ClassicSimilarity], result of:
           0.17542592 = score(doc=389,freq=2.0), product of:
             0.27039093 = queryWeight, product of:
               2.7323132 = boost
               7.3401785 = idf(docFreq=77, maxDocs=44218)
               0.013482022 = queryNorm
             0.64878625 = fieldWeight in 389, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               7.3401785 = idf(docFreq=77, maxDocs=44218)
               0.0625 = fieldNorm(doc=389)
         0.15034825 = weight(abstract_txt:citation in 389) [ClassicSimilarity], result of:
           0.15034825 = score(doc=389,freq=6.0), product of:
             0.20055693 = queryWeight, product of:
               3.037929 = boost
               4.896717 = idf(docFreq=897, maxDocs=44218)
               0.013482022 = queryNorm
             0.7496537 = fieldWeight in 389, product of:
               2.4494898 = tf(freq=6.0), with freq of:
                 6.0 = termFreq=6.0
               4.896717 = idf(docFreq=897, maxDocs=44218)
               0.0625 = fieldNorm(doc=389)
       0.24 = coord(6/25)