Document (#41222)

Author

Çelebi, A.
Özgür, A.

Title

Segmenting hashtags and analyzing their grammatical structure

Source

Journal of the Association for Information Science and Technology. 69(2018) no.5, S.675-686

Year

2018

Abstract

Originated as a label to mark specific tweets, hashtags are increasingly used to convey messages that people like to see in the trending hashtags list. Complex noun phrases and even sentences can be turned into a hashtag. Breaking hashtags into their words is a challenging task due to the irregular and compact nature of the language used in Twitter. In this study, we investigate feature-based machine learning and language model (LM)-based approaches for hashtag segmentation. Our results show that LM alone is not successful at segmenting nontrivial hashtags. However, when the N-best LM-based segmentations are incorporated as features into the feature-based approach, along with context-based features proposed in this study, state-of-the-art results in hashtag segmentation are achieved. In addition, we provide an analysis of over two million distinct hashtags, autosegmented by using our best configuration. The analysis reveals that half of all 60 million hashtag occurrences contain multiple words and 80% of sentiment is trapped inside multiword hashtags, justifying the need for hashtag segmentation. Furthermore, we analyze the grammatical structure of hashtags by parsing them and observe that 77% of the hashtags are noun-based, whereas 11.9% are verb-based.

Content

Vgl.: https://onlinelibrary.wiley.com/doi/abs/10.1002/asi.23989.

Theme

Metadaten
Computerlinguistik

Object

Hashtag

Similar documents (content)

1. Ma, Z.; Sun, A.; Cong, G.: On predicting the popularity of newly emerging hashtags in Twitter (2013) 0.39

   0.387488 = sum of:
     0.387488 = product of:
       1.3838857 = sum of:
         0.032056876 = weight(abstract_txt:features in 1967) [ClassicSimilarity], result of:
           0.032056876 = score(doc=1967,freq=7.0), product of:
             0.042694982 = queryWeight, product of:
               1.1261628 = boost
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.008349498 = queryNorm
             0.7508347 = fieldWeight in 1967, product of:
               2.6457512 = tf(freq=7.0), with freq of:
                 7.0 = termFreq=7.0
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.0625 = fieldNorm(doc=1967)
         0.005930271 = weight(abstract_txt:that in 1967) [ClassicSimilarity], result of:
           0.005930271 = score(doc=1967,freq=3.0), product of:
             0.02316406 = queryWeight, product of:
               1.1731001 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.008349498 = queryNorm
             0.25601172 = fieldWeight in 1967, 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=1967)
         0.016342595 = weight(abstract_txt:best in 1967) [ClassicSimilarity], result of:
           0.016342595 = score(doc=1967,freq=1.0), product of:
             0.0521206 = queryWeight, product of:
               1.2442781 = boost
               5.0168557 = idf(docFreq=799, maxDocs=44421)
               0.008349498 = queryNorm
             0.31355348 = fieldWeight in 1967, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.0168557 = idf(docFreq=799, maxDocs=44421)
               0.0625 = fieldNorm(doc=1967)
         0.04380075 = weight(abstract_txt:million in 1967) [ClassicSimilarity], result of:
           0.04380075 = score(doc=1967,freq=2.0), product of:
             0.07981902 = queryWeight, product of:
               1.5398058 = boost
               6.208406 = idf(docFreq=242, maxDocs=44421)
               0.008349498 = queryNorm
             0.54875076 = fieldWeight in 1967, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               6.208406 = idf(docFreq=242, maxDocs=44421)
               0.0625 = fieldNorm(doc=1967)
         0.014609487 = weight(abstract_txt:based in 1967) [ClassicSimilarity], result of:
           0.014609487 = score(doc=1967,freq=1.0), product of:
             0.073435836 = queryWeight, product of:
               2.7631266 = boost
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.008349498 = queryNorm
             0.1989422 = fieldWeight in 1967, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.0625 = fieldNorm(doc=1967)
         0.520335 = weight(abstract_txt:hashtag in 1967) [ClassicSimilarity], result of:
           0.520335 = score(doc=1967,freq=3.0), product of:
             0.49270883 = queryWeight, product of:
               6.048921 = boost
               9.755557 = idf(docFreq=6, maxDocs=44421)
               0.008349498 = queryNorm
             1.05607 = fieldWeight in 1967, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               9.755557 = idf(docFreq=6, maxDocs=44421)
               0.0625 = fieldNorm(doc=1967)
         0.75081074 = weight(abstract_txt:hashtags in 1967) [ClassicSimilarity], result of:
           0.75081074 = score(doc=1967,freq=3.0), product of:
             0.76532537 = queryWeight, product of:
               10.114445 = boost
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.008349498 = queryNorm
             0.9810347 = fieldWeight in 1967, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.0625 = fieldNorm(doc=1967)
       0.28 = coord(7/25)
   

2. Kong, S.; Ye, F.; Feng, L.; Zhao, Z.: Towards the prediction problems of bursting hashtags on Twitter (2015) 0.34

   0.3376549 = sum of:
     0.3376549 = product of:
       1.6882744 = sum of:
         0.02570268 = weight(abstract_txt:features in 3338) [ClassicSimilarity], result of:
           0.02570268 = score(doc=3338,freq=2.0), product of:
             0.042694982 = queryWeight, product of:
               1.1261628 = boost
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.008349498 = queryNorm
             0.60200703 = fieldWeight in 3338, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.09375 = fieldNorm(doc=3338)
         0.08059649 = weight(abstract_txt:trending in 3338) [ClassicSimilarity], result of:
           0.08059649 = score(doc=3338,freq=1.0), product of:
             0.09146789 = queryWeight, product of:
               1.1655531 = boost
               9.398883 = idf(docFreq=9, maxDocs=44421)
               0.008349498 = queryNorm
             0.88114524 = fieldWeight in 3338, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.398883 = idf(docFreq=9, maxDocs=44421)
               0.09375 = fieldNorm(doc=3338)
         0.0051357658 = weight(abstract_txt:that in 3338) [ClassicSimilarity], result of:
           0.0051357658 = score(doc=3338,freq=1.0), product of:
             0.02316406 = queryWeight, product of:
               1.1731001 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.008349498 = queryNorm
             0.22171268 = fieldWeight in 3338, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.09375 = fieldNorm(doc=3338)
         0.45062333 = weight(abstract_txt:hashtag in 3338) [ClassicSimilarity], result of:
           0.45062333 = score(doc=3338,freq=1.0), product of:
             0.49270883 = queryWeight, product of:
               6.048921 = boost
               9.755557 = idf(docFreq=6, maxDocs=44421)
               0.008349498 = queryNorm
             0.91458344 = fieldWeight in 3338, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.755557 = idf(docFreq=6, maxDocs=44421)
               0.09375 = fieldNorm(doc=3338)
         1.126216 = weight(abstract_txt:hashtags in 3338) [ClassicSimilarity], result of:
           1.126216 = score(doc=3338,freq=3.0), product of:
             0.76532537 = queryWeight, product of:
               10.114445 = boost
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.008349498 = queryNorm
             1.471552 = fieldWeight in 3338, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.09375 = fieldNorm(doc=3338)
       0.2 = coord(5/25)
   

3. Chang, H.-C.; Iyer, I.: Trends in Twitter hashtag applications : design features for value-added dimensions to future library catalogues (2012) 0.25

   0.24811828 = sum of:
     0.24811828 = product of:
       1.5507393 = sum of:
         0.015145451 = weight(abstract_txt:features in 574) [ClassicSimilarity], result of:
           0.015145451 = score(doc=574,freq=1.0), product of:
             0.042694982 = queryWeight, product of:
               1.1261628 = boost
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.008349498 = queryNorm
             0.3547361 = fieldWeight in 574, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.078125 = fieldNorm(doc=574)
         0.018261857 = weight(abstract_txt:based in 574) [ClassicSimilarity], result of:
           0.018261857 = score(doc=574,freq=1.0), product of:
             0.073435836 = queryWeight, product of:
               2.7631266 = boost
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.008349498 = queryNorm
             0.24867775 = fieldWeight in 574, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.078125 = fieldNorm(doc=574)
         0.75103897 = weight(abstract_txt:hashtag in 574) [ClassicSimilarity], result of:
           0.75103897 = score(doc=574,freq=4.0), product of:
             0.49270883 = queryWeight, product of:
               6.048921 = boost
               9.755557 = idf(docFreq=6, maxDocs=44421)
               0.008349498 = queryNorm
             1.5243058 = fieldWeight in 574, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               9.755557 = idf(docFreq=6, maxDocs=44421)
               0.078125 = fieldNorm(doc=574)
         0.766293 = weight(abstract_txt:hashtags in 574) [ClassicSimilarity], result of:
           0.766293 = score(doc=574,freq=2.0), product of:
             0.76532537 = queryWeight, product of:
               10.114445 = boost
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.008349498 = queryNorm
             1.0012643 = fieldWeight in 574, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.078125 = fieldNorm(doc=574)
       0.16 = coord(4/25)
   

4. Sedhai, S.; Sun, A.: ¬An analysis of 14 Million tweets on hashtag-oriented spamming* (2017) 0.21

   0.21248916 = sum of:
     0.21248916 = product of:
       1.0624458 = sum of:
         0.004842046 = weight(abstract_txt:that in 4683) [ClassicSimilarity], result of:
           0.004842046 = score(doc=4683,freq=2.0), product of:
             0.02316406 = queryWeight, product of:
               1.1731001 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.008349498 = queryNorm
             0.20903271 = fieldWeight in 4683, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.0625 = fieldNorm(doc=4683)
         0.028120423 = weight(abstract_txt:words in 4683) [ClassicSimilarity], result of:
           0.028120423 = score(doc=4683,freq=2.0), product of:
             0.059401847 = queryWeight, product of:
               1.3283505 = boost
               5.355831 = idf(docFreq=569, maxDocs=44421)
               0.008349498 = queryNorm
             0.47339305 = fieldWeight in 4683, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               5.355831 = idf(docFreq=569, maxDocs=44421)
               0.0625 = fieldNorm(doc=4683)
         0.030971806 = weight(abstract_txt:million in 4683) [ClassicSimilarity], result of:
           0.030971806 = score(doc=4683,freq=1.0), product of:
             0.07981902 = queryWeight, product of:
               1.5398058 = boost
               6.208406 = idf(docFreq=242, maxDocs=44421)
               0.008349498 = queryNorm
             0.38802537 = fieldWeight in 4683, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               6.208406 = idf(docFreq=242, maxDocs=44421)
               0.0625 = fieldNorm(doc=4683)
         0.029218974 = weight(abstract_txt:based in 4683) [ClassicSimilarity], result of:
           0.029218974 = score(doc=4683,freq=4.0), product of:
             0.073435836 = queryWeight, product of:
               2.7631266 = boost
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.008349498 = queryNorm
             0.3978844 = fieldWeight in 4683, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.0625 = fieldNorm(doc=4683)
         0.9692925 = weight(abstract_txt:hashtags in 4683) [ClassicSimilarity], result of:
           0.9692925 = score(doc=4683,freq=5.0), product of:
             0.76532537 = queryWeight, product of:
               10.114445 = boost
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.008349498 = queryNorm
             1.2665104 = fieldWeight in 4683, product of:
               2.236068 = tf(freq=5.0), with freq of:
                 5.0 = termFreq=5.0
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.0625 = fieldNorm(doc=4683)
       0.2 = coord(5/25)
   

5. Luo, Z.; Yu, Y.; Osborne, M.; Wang, T.: Structuring tweets for improving Twitter search (2015) 0.15

   0.145122 = sum of:
     0.145122 = product of:
       0.5182929 = sum of:
         0.010712732 = weight(abstract_txt:structure in 3335) [ClassicSimilarity], result of:
           0.010712732 = score(doc=3335,freq=1.0), product of:
             0.039330445 = queryWeight, product of:
               1.0808792 = boost
               4.3580413 = idf(docFreq=1545, maxDocs=44421)
               0.008349498 = queryNorm
             0.27237758 = fieldWeight in 3335, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.3580413 = idf(docFreq=1545, maxDocs=44421)
               0.0625 = fieldNorm(doc=3335)
         0.01713512 = weight(abstract_txt:features in 3335) [ClassicSimilarity], result of:
           0.01713512 = score(doc=3335,freq=2.0), product of:
             0.042694982 = queryWeight, product of:
               1.1261628 = boost
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.008349498 = queryNorm
             0.40133804 = fieldWeight in 3335, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.5406218 = idf(docFreq=1287, maxDocs=44421)
               0.0625 = fieldNorm(doc=3335)
         0.005930271 = weight(abstract_txt:that in 3335) [ClassicSimilarity], result of:
           0.005930271 = score(doc=3335,freq=3.0), product of:
             0.02316406 = queryWeight, product of:
               1.1731001 = boost
               2.3649352 = idf(docFreq=11344, maxDocs=44421)
               0.008349498 = queryNorm
             0.25601172 = fieldWeight in 3335, 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=3335)
         0.009810741 = weight(abstract_txt:into in 3335) [ClassicSimilarity], result of:
           0.009810741 = score(doc=3335,freq=1.0), product of:
             0.04245808 = queryWeight, product of:
               1.3754302 = boost
               3.697102 = idf(docFreq=2993, maxDocs=44421)
               0.008349498 = queryNorm
             0.23106888 = fieldWeight in 3335, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.697102 = idf(docFreq=2993, maxDocs=44421)
               0.0625 = fieldNorm(doc=3335)
         0.026613781 = weight(abstract_txt:feature in 3335) [ClassicSimilarity], result of:
           0.026613781 = score(doc=3335,freq=1.0), product of:
             0.07214393 = queryWeight, product of:
               1.463904 = boost
               5.9023747 = idf(docFreq=329, maxDocs=44421)
               0.008349498 = queryNorm
             0.36889842 = fieldWeight in 3335, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.9023747 = idf(docFreq=329, maxDocs=44421)
               0.0625 = fieldNorm(doc=3335)
         0.014609487 = weight(abstract_txt:based in 3335) [ClassicSimilarity], result of:
           0.014609487 = score(doc=3335,freq=1.0), product of:
             0.073435836 = queryWeight, product of:
               2.7631266 = boost
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.008349498 = queryNorm
             0.1989422 = fieldWeight in 3335, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.1830752 = idf(docFreq=5005, maxDocs=44421)
               0.0625 = fieldNorm(doc=3335)
         0.43348077 = weight(abstract_txt:hashtags in 3335) [ClassicSimilarity], result of:
           0.43348077 = score(doc=3335,freq=1.0), product of:
             0.76532537 = queryWeight, product of:
               10.114445 = boost
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.008349498 = queryNorm
             0.56640065 = fieldWeight in 3335, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               9.06241 = idf(docFreq=13, maxDocs=44421)
               0.0625 = fieldNorm(doc=3335)
       0.28 = coord(7/25)