Document (#36243)

Author

Girju, R.
Beamer, B.
Rozovskaya, A.
Fister, A.
Bhat, S.

Title

¬A knowledge-rich approach to identifying semantic relations between nominals

Source

Information processing and management. 46(2010) no.5, S.589-610

Year

2010

Abstract

This paper describes a state-of-the-art supervised, knowledge-intensive approach to the automatic identification of semantic relations between nominals in English sentences. The system employs a combination of rich and varied sets of new and previously used lexical, syntactic, and semantic features extracted from various knowledge sources such as WordNet and additional annotated corpora. The system ranked first at the third most popular SemEval 2007 Task - Classification of Semantic Relations between Nominals and achieved an F-measure of 72.4% and an accuracy of 76.3%. We also show that some semantic relations are better suited for WordNet-based models than other relations. Additionally, we make a distinction between out-of-context (regular) examples and those that require sentence context for relation identification and show that contextual data are important for the performance of a noun-noun semantic parser. Finally, learning curves show that the task difficulty varies across relations and that our learned WordNet-based representation is highly accurate so the performance results suggest the upper bound on what this representation can do.

Theme

Wissensrepräsentation

Similar documents (content)

1. Leroy, G.; Chen, H.: Genescene: an ontology-enhanced integration of linguistic and co-occurrence based relations in biomedical texts (2005) 0.36

   0.35861453 = sum of:
     0.35861453 = product of:
       0.99615145 = sum of:
         0.156857 = weight(abstract_txt:parser in 259) [ClassicSimilarity], result of:
           0.156857 = score(doc=259,freq=3.0), product of:
             0.17457058 = queryWeight, product of:
               1.1824864 = boost
               8.30027 = idf(docFreq=29, maxDocs=44421)
               0.017786179 = queryNorm
             0.8985306 = fieldWeight in 259, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               8.30027 = idf(docFreq=29, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.02119085 = weight(abstract_txt:knowledge in 259) [ClassicSimilarity], result of:
           0.02119085 = score(doc=259,freq=1.0), product of:
             0.0956052 = queryWeight, product of:
               1.5156958 = boost
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.017786179 = queryNorm
             0.22164954 = fieldWeight in 259, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.018140739 = weight(abstract_txt:that in 259) [ClassicSimilarity], result of:
           0.018140739 = score(doc=259,freq=3.0), product of:
             0.070859015 = queryWeight, product of:
               1.684586 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.017786179 = queryNorm
             0.25601172 = fieldWeight in 259, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.07109857 = weight(abstract_txt:rich in 259) [ClassicSimilarity], result of:
           0.07109857 = score(doc=259,freq=1.0), product of:
             0.18718 = queryWeight, product of:
               1.731631 = boost
               6.0774503 = idf(docFreq=276, maxDocs=44421)
               0.017786179 = queryNorm
             0.37984064 = fieldWeight in 259, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               6.0774503 = idf(docFreq=276, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.040498637 = weight(abstract_txt:show in 259) [ClassicSimilarity], result of:
           0.040498637 = score(doc=259,freq=1.0), product of:
             0.14723448 = queryWeight, product of:
               1.8809447 = boost
               4.400995 = idf(docFreq=1480, maxDocs=44421)
               0.017786179 = queryNorm
             0.27506217 = fieldWeight in 259, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.400995 = idf(docFreq=1480, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.02618027 = weight(abstract_txt:between in 259) [ClassicSimilarity], result of:
           0.02618027 = score(doc=259,freq=1.0), product of:
             0.121156 = queryWeight, product of:
               1.9702127 = boost
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.017786179 = queryNorm
             0.21608727 = fieldWeight in 259, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.14855923 = weight(abstract_txt:noun in 259) [ClassicSimilarity], result of:
           0.14855923 = score(doc=259,freq=1.0), product of:
             0.30592757 = queryWeight, product of:
               2.2137825 = boost
               7.769642 = idf(docFreq=50, maxDocs=44421)
               0.017786179 = queryNorm
             0.48560262 = fieldWeight in 259, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.769642 = idf(docFreq=50, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.08512569 = weight(abstract_txt:semantic in 259) [ClassicSimilarity], result of:
           0.08512569 = score(doc=259,freq=1.0), product of:
             0.30439186 = queryWeight, product of:
               3.8247478 = boost
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.017786179 = queryNorm
             0.27965823 = fieldWeight in 259, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
         0.42850047 = weight(abstract_txt:relations in 259) [ClassicSimilarity], result of:
           0.42850047 = score(doc=259,freq=7.0), product of:
             0.46736997 = queryWeight, product of:
               4.739327 = boost
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.017786179 = queryNorm
             0.9168336 = fieldWeight in 259, product of:
               2.6457512 = tf(freq=7.0), with freq of:
                 7.0 = termFreq=7.0
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.0625 = fieldNorm(doc=259)
       0.36 = coord(9/25)
   

2. Jouis, C.: System of types + inter-concept relations properties : towards validation of constructed terminologies (1998) 0.29

   0.28560165 = sum of:
     0.28560165 = product of:
       0.89250517 = sum of:
         0.04674401 = weight(abstract_txt:task in 1056) [ClassicSimilarity], result of:
           0.04674401 = score(doc=1056,freq=1.0), product of:
             0.1219634 = queryWeight, product of:
               1.3977851 = boost
               4.9057617 = idf(docFreq=893, maxDocs=44421)
               0.017786179 = queryNorm
             0.38326263 = fieldWeight in 1056, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.9057617 = idf(docFreq=893, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
         0.06693168 = weight(abstract_txt:representation in 1056) [ClassicSimilarity], result of:
           0.06693168 = score(doc=1056,freq=2.0), product of:
             0.12297685 = queryWeight, product of:
               1.4035805 = boost
               4.9261017 = idf(docFreq=875, maxDocs=44421)
               0.017786179 = queryNorm
             0.54426247 = fieldWeight in 1056, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.9261017 = idf(docFreq=875, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
         0.045879535 = weight(abstract_txt:knowledge in 1056) [ClassicSimilarity], result of:
           0.045879535 = score(doc=1056,freq=3.0), product of:
             0.0956052 = queryWeight, product of:
               1.5156958 = boost
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.017786179 = queryNorm
             0.47988534 = fieldWeight in 1056, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
         0.078541145 = weight(abstract_txt:identification in 1056) [ClassicSimilarity], result of:
           0.078541145 = score(doc=1056,freq=1.0), product of:
             0.17237608 = queryWeight, product of:
               1.6617441 = boost
               5.8321705 = idf(docFreq=353, maxDocs=44421)
               0.017786179 = queryNorm
             0.45563832 = fieldWeight in 1056, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.8321705 = idf(docFreq=353, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
         0.018514816 = weight(abstract_txt:that in 1056) [ClassicSimilarity], result of:
           0.018514816 = score(doc=1056,freq=2.0), product of:
             0.070859015 = queryWeight, product of:
               1.684586 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.017786179 = queryNorm
             0.2612909 = fieldWeight in 1056, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
         0.032725338 = weight(abstract_txt:between in 1056) [ClassicSimilarity], result of:
           0.032725338 = score(doc=1056,freq=1.0), product of:
             0.121156 = queryWeight, product of:
               1.9702127 = boost
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.017786179 = queryNorm
             0.2701091 = fieldWeight in 1056, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
         0.15048239 = weight(abstract_txt:semantic in 1056) [ClassicSimilarity], result of:
           0.15048239 = score(doc=1056,freq=2.0), product of:
             0.30439186 = queryWeight, product of:
               3.8247478 = boost
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.017786179 = queryNorm
             0.49437058 = fieldWeight in 1056, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
         0.45268628 = weight(abstract_txt:relations in 1056) [ClassicSimilarity], result of:
           0.45268628 = score(doc=1056,freq=5.0), product of:
             0.46736997 = queryWeight, product of:
               4.739327 = boost
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.017786179 = queryNorm
             0.9685823 = fieldWeight in 1056, product of:
               2.236068 = tf(freq=5.0), with freq of:
                 5.0 = termFreq=5.0
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.078125 = fieldNorm(doc=1056)
       0.32 = coord(8/25)
   

3. Blanco, E.; Cankaya, H.C.; Moldovan, D.: Composition of semantic relations : model and applications (2010) 0.25

   0.25233325 = sum of:
     0.25233325 = product of:
       1.2616663 = sum of:
         0.1584825 = weight(abstract_txt:parser in 761) [ClassicSimilarity], result of:
           0.1584825 = score(doc=761,freq=1.0), product of:
             0.17457058 = queryWeight, product of:
               1.1824864 = boost
               8.30027 = idf(docFreq=29, maxDocs=44421)
               0.017786179 = queryNorm
             0.90784204 = fieldWeight in 761, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               8.30027 = idf(docFreq=29, maxDocs=44421)
               0.109375 = fieldNorm(doc=761)
         0.018328732 = weight(abstract_txt:that in 761) [ClassicSimilarity], result of:
           0.018328732 = score(doc=761,freq=1.0), product of:
             0.070859015 = queryWeight, product of:
               1.684586 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.017786179 = queryNorm
             0.2586648 = fieldWeight in 761, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.109375 = fieldNorm(doc=761)
         0.25997865 = weight(abstract_txt:noun in 761) [ClassicSimilarity], result of:
           0.25997865 = score(doc=761,freq=1.0), product of:
             0.30592757 = queryWeight, product of:
               2.2137825 = boost
               7.769642 = idf(docFreq=50, maxDocs=44421)
               0.017786179 = queryNorm
             0.8498046 = fieldWeight in 761, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.769642 = idf(docFreq=50, maxDocs=44421)
               0.109375 = fieldNorm(doc=761)
         0.25802353 = weight(abstract_txt:semantic in 761) [ClassicSimilarity], result of:
           0.25802353 = score(doc=761,freq=3.0), product of:
             0.30439186 = queryWeight, product of:
               3.8247478 = boost
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.017786179 = queryNorm
             0.84766895 = fieldWeight in 761, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.109375 = fieldNorm(doc=761)
         0.56685287 = weight(abstract_txt:relations in 761) [ClassicSimilarity], result of:
           0.56685287 = score(doc=761,freq=4.0), product of:
             0.46736997 = queryWeight, product of:
               4.739327 = boost
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.017786179 = queryNorm
             1.2128569 = fieldWeight in 761, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.109375 = fieldNorm(doc=761)
       0.2 = coord(5/25)
   

4. Khoo, S.G.; Na, J.-C.: Semantic relations in information science (2006) 0.23

   0.22951879 = sum of:
     0.22951879 = product of:
       0.71724623 = sum of:
         0.029995283 = weight(abstract_txt:regular in 2978) [ClassicSimilarity], result of:
           0.029995283 = score(doc=2978,freq=1.0), product of:
             0.1326572 = queryWeight, product of:
               1.0308039 = boost
               7.2355595 = idf(docFreq=86, maxDocs=44421)
               0.017786179 = queryNorm
             0.22611123 = fieldWeight in 2978, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.2355595 = idf(docFreq=86, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
         0.01893114 = weight(abstract_txt:representation in 2978) [ClassicSimilarity], result of:
           0.01893114 = score(doc=2978,freq=1.0), product of:
             0.12297685 = queryWeight, product of:
               1.4035805 = boost
               4.9261017 = idf(docFreq=875, maxDocs=44421)
               0.017786179 = queryNorm
             0.15394068 = fieldWeight in 2978, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.9261017 = idf(docFreq=875, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
         0.028032858 = weight(abstract_txt:knowledge in 2978) [ClassicSimilarity], result of:
           0.028032858 = score(doc=2978,freq=7.0), product of:
             0.0956052 = queryWeight, product of:
               1.5156958 = boost
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.017786179 = queryNorm
             0.29321477 = fieldWeight in 2978, product of:
               2.6457512 = tf(freq=7.0), with freq of:
                 7.0 = termFreq=7.0
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
         0.03141646 = weight(abstract_txt:identification in 2978) [ClassicSimilarity], result of:
           0.03141646 = score(doc=2978,freq=1.0), product of:
             0.17237608 = queryWeight, product of:
               1.6617441 = boost
               5.8321705 = idf(docFreq=353, maxDocs=44421)
               0.017786179 = queryNorm
             0.18225533 = fieldWeight in 2978, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.8321705 = idf(docFreq=353, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
         0.01959424 = weight(abstract_txt:that in 2978) [ClassicSimilarity], result of:
           0.01959424 = score(doc=2978,freq=14.0), product of:
             0.070859015 = queryWeight, product of:
               1.684586 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.017786179 = queryNorm
             0.2765243 = fieldWeight in 2978, product of:
               3.7416575 = tf(freq=14.0), with freq of:
                 14.0 = termFreq=14.0
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
         0.02618027 = weight(abstract_txt:between in 2978) [ClassicSimilarity], result of:
           0.02618027 = score(doc=2978,freq=4.0), product of:
             0.121156 = queryWeight, product of:
               1.9702127 = boost
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.017786179 = queryNorm
             0.21608727 = fieldWeight in 2978, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
         0.13459553 = weight(abstract_txt:semantic in 2978) [ClassicSimilarity], result of:
           0.13459553 = score(doc=2978,freq=10.0), product of:
             0.30439186 = queryWeight, product of:
               3.8247478 = boost
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.017786179 = queryNorm
             0.4421785 = fieldWeight in 2978, product of:
               3.1622777 = tf(freq=10.0), with freq of:
                 10.0 = termFreq=10.0
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
         0.42850047 = weight(abstract_txt:relations in 2978) [ClassicSimilarity], result of:
           0.42850047 = score(doc=2978,freq=28.0), product of:
             0.46736997 = queryWeight, product of:
               4.739327 = boost
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.017786179 = queryNorm
             0.9168336 = fieldWeight in 2978, product of:
               5.2915025 = tf(freq=28.0), with freq of:
                 28.0 = termFreq=28.0
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.03125 = fieldNorm(doc=2978)
       0.32 = coord(8/25)
   

5. Murphy, M.L.: Semantic relations and the lexicon : antonymy, synonymy and other paradigms (2008) 0.23

   0.22844723 = sum of:
     0.22844723 = product of:
       0.815883 = sum of:
         0.029318033 = weight(abstract_txt:approach in 1997) [ClassicSimilarity], result of:
           0.029318033 = score(doc=1997,freq=2.0), product of:
             0.07092926 = queryWeight, product of:
               1.0659537 = boost
               3.741144 = idf(docFreq=2864, maxDocs=44421)
               0.017786179 = queryNorm
             0.41334188 = fieldWeight in 1997, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               3.741144 = idf(docFreq=2864, maxDocs=44421)
               0.078125 = fieldNorm(doc=1997)
         0.047327846 = weight(abstract_txt:representation in 1997) [ClassicSimilarity], result of:
           0.047327846 = score(doc=1997,freq=1.0), product of:
             0.12297685 = queryWeight, product of:
               1.4035805 = boost
               4.9261017 = idf(docFreq=875, maxDocs=44421)
               0.017786179 = queryNorm
             0.3848517 = fieldWeight in 1997, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.9261017 = idf(docFreq=875, maxDocs=44421)
               0.078125 = fieldNorm(doc=1997)
         0.03746048 = weight(abstract_txt:knowledge in 1997) [ClassicSimilarity], result of:
           0.03746048 = score(doc=1997,freq=2.0), product of:
             0.0956052 = queryWeight, product of:
               1.5156958 = boost
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.017786179 = queryNorm
             0.39182472 = fieldWeight in 1997, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               3.5463927 = idf(docFreq=3480, maxDocs=44421)
               0.078125 = fieldNorm(doc=1997)
         0.022675924 = weight(abstract_txt:that in 1997) [ClassicSimilarity], result of:
           0.022675924 = score(doc=1997,freq=3.0), product of:
             0.070859015 = queryWeight, product of:
               1.684586 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.017786179 = queryNorm
             0.32001466 = fieldWeight in 1997, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.078125 = fieldNorm(doc=1997)
         0.032725338 = weight(abstract_txt:between in 1997) [ClassicSimilarity], result of:
           0.032725338 = score(doc=1997,freq=1.0), product of:
             0.121156 = queryWeight, product of:
               1.9702127 = boost
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.017786179 = queryNorm
             0.2701091 = fieldWeight in 1997, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.4573963 = idf(docFreq=3804, maxDocs=44421)
               0.078125 = fieldNorm(doc=1997)
         0.15048239 = weight(abstract_txt:semantic in 1997) [ClassicSimilarity], result of:
           0.15048239 = score(doc=1997,freq=2.0), product of:
             0.30439186 = queryWeight, product of:
               3.8247478 = boost
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.017786179 = queryNorm
             0.49437058 = fieldWeight in 1997, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.4745317 = idf(docFreq=1375, maxDocs=44421)
               0.078125 = fieldNorm(doc=1997)
         0.49589297 = weight(abstract_txt:relations in 1997) [ClassicSimilarity], result of:
           0.49589297 = score(doc=1997,freq=6.0), product of:
             0.46736997 = queryWeight, product of:
               4.739327 = boost
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.017786179 = queryNorm
             1.0610287 = fieldWeight in 1997, product of:
               2.4494898 = tf(freq=6.0), with freq of:
                 6.0 = termFreq=6.0
               5.5444884 = idf(docFreq=471, maxDocs=44421)
               0.078125 = fieldNorm(doc=1997)
       0.28 = coord(7/25)