Document (#44020)

Author

Liu, J.
Zhou, Z.
Gao, M.
Tang, J.
Fan, W.

Title

Aspect sentiment mining of short bullet screen comments from online TV series

Source

Journal of the Association for Information Science and Technology. 74(2023) no.8, S.1026-1045

Year

2023

Abstract

Bullet screen comments (BSCs) are user-generated short comments that appear as real-time overlays on many video platforms, expressing the audience opinions and emotions about different aspects of the ongoing video. Unlike traditional long comments after a show, BSCs are often incomplete, ambiguous in context, and correlated over time. Current studies in sentiment analysis of BSCs rarely address these challenges, motivating us to develop an aspect-level sentiment analysis framework. Our framework, BSCNET, is a pre-trained language encoder-based deep neural classifier designed to enhance semantic understanding. A novel neighbor context construction method is proposed to uncover latent contextual correlation among BSCs over time, and we also incorporate semi-supervised learning to reduce labeling costs. The framework increases F1 (Macro) and accuracy by up to 10% and 10.2%, respectively. Additionally, we have developed two novel downstream tasks. The first is noisy BSCs identification, which reached F1 (Macro) and accuracy of 90.1% and 98.3%, respectively, through fine-tuning the BSCNET. The second is the prediction of future episode popularity, where the MAPE is reduced by 11%-19.0% when incorporating sentiment features. Overall, this study provides a methodology reference for aspect-level sentiment analysis of BSCs and highlights its potential for viewing experience or forthcoming content optimization.

Content

Vgl.: https://asistdl.onlinelibrary.wiley.com/doi/10.1002/asi.24800.

Similar documents (author)

1. Zhou, Z.; Jin, X.-L.; Hsu, C.; Tang, Z.: User empowerment and well-being with mHealth apps during pandemics : a mix-methods investigation in China (2023) 3.49

   3.4933286 = sum of:
     3.4933286 = sum of:
       1.6944311 = weight(author_txt:zhou in 1090) [ClassicSimilarity], result of:
         1.6944311 = score(doc=1090,freq=1.0), product of:
           0.69286954 = queryWeight, product of:
             7.825686 = idf(docFreq=47, maxDocs=44218)
             0.088537864 = queryNorm
           2.4455268 = fieldWeight in 1090, product of:
             1.0 = tf(freq=1.0), with freq of:
               1.0 = termFreq=1.0
             7.825686 = idf(docFreq=47, maxDocs=44218)
             0.3125 = fieldNorm(doc=1090)
       1.7988976 = weight(author_txt:tang in 1090) [ClassicSimilarity], result of:
         1.7988976 = score(doc=1090,freq=1.0), product of:
           0.7210629 = queryWeight, product of:
             1.0201426 = boost
             7.983315 = idf(docFreq=40, maxDocs=44218)
             0.088537864 = queryNorm
           2.494786 = fieldWeight in 1090, product of:
             1.0 = tf(freq=1.0), with freq of:
               1.0 = termFreq=1.0
             7.983315 = idf(docFreq=40, maxDocs=44218)
             0.3125 = fieldNorm(doc=1090)
   

2. Zhou, L.: Characteristics of material organization and classification in the Kinsey Institute Library (2003) 1.69

   1.6944311 = sum of:
     1.6944311 = product of:
       3.3888621 = sum of:
         3.3888621 = weight(author_txt:zhou in 5639) [ClassicSimilarity], result of:
           3.3888621 = score(doc=5639,freq=1.0), product of:
             0.69286954 = queryWeight, product of:
               7.825686 = idf(docFreq=47, maxDocs=44218)
               0.088537864 = queryNorm
             4.8910537 = fieldWeight in 5639, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.825686 = idf(docFreq=47, maxDocs=44218)
               0.625 = fieldNorm(doc=5639)
       0.5 = coord(1/2)
   

3. Tang, R.; Solomon, P.: Toward an understanding of the dynamics of relevance judgement : an analysis of one person's search behavior (1998) 1.44

   1.439118 = sum of:
     1.439118 = product of:
       2.878236 = sum of:
         2.878236 = weight(author_txt:tang in 3268) [ClassicSimilarity], result of:
           2.878236 = score(doc=3268,freq=1.0), product of:
             0.7210629 = queryWeight, product of:
               1.0201426 = boost
               7.983315 = idf(docFreq=40, maxDocs=44218)
               0.088537864 = queryNorm
             3.9916575 = fieldWeight in 3268, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.983315 = idf(docFreq=40, maxDocs=44218)
               0.5 = fieldNorm(doc=3268)
       0.5 = coord(1/2)
   

4. Tang, R.; Solomon, P.: Use of relevance criteria across stages of document evaluation : on the complementarity of experimental and naturalistic studies (2001) 1.44

   1.439118 = sum of:
     1.439118 = product of:
       2.878236 = sum of:
         2.878236 = weight(author_txt:tang in 5213) [ClassicSimilarity], result of:
           2.878236 = score(doc=5213,freq=1.0), product of:
             0.7210629 = queryWeight, product of:
               1.0201426 = boost
               7.983315 = idf(docFreq=40, maxDocs=44218)
               0.088537864 = queryNorm
             3.9916575 = fieldWeight in 5213, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.983315 = idf(docFreq=40, maxDocs=44218)
               0.5 = fieldNorm(doc=5213)
       0.5 = coord(1/2)
   

5. Tang, M.-C.: Browsing and searching in a faceted information space : a naturalistic study of PubMed users' interaction with a display tool (2007) 1.44

   1.439118 = sum of:
     1.439118 = product of:
       2.878236 = sum of:
         2.878236 = weight(author_txt:tang in 617) [ClassicSimilarity], result of:
           2.878236 = score(doc=617,freq=1.0), product of:
             0.7210629 = queryWeight, product of:
               1.0201426 = boost
               7.983315 = idf(docFreq=40, maxDocs=44218)
               0.088537864 = queryNorm
             3.9916575 = fieldWeight in 617, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               7.983315 = idf(docFreq=40, maxDocs=44218)
               0.5 = fieldNorm(doc=617)
       0.5 = coord(1/2)
   

Similar documents (content)

1. Chen, Z.; Huang, Y.; Tian, J.; Liu, X.; Fu, K.; Huang, T.: Joint model for subsentence-level sentiment analysis with Markov logic (2015) 0.23

   0.2269919 = sum of:
     0.2269919 = product of:
       0.94579965 = sum of:
         0.03071266 = weight(abstract_txt:level in 2210) [ClassicSimilarity], result of:
           0.03071266 = score(doc=2210,freq=2.0), product of:
             0.07725155 = queryWeight, product of:
               1.0754635 = boost
               4.497956 = idf(docFreq=1337, maxDocs=44218)
               0.015969684 = queryNorm
             0.39756688 = fieldWeight in 2210, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.497956 = idf(docFreq=1337, maxDocs=44218)
               0.0625 = fieldNorm(doc=2210)
         0.0427629 = weight(abstract_txt:analysis in 2210) [ClassicSimilarity], result of:
           0.0427629 = score(doc=2210,freq=6.0), product of:
             0.07645344 = queryWeight, product of:
               1.3103467 = boost
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.015969684 = queryNorm
             0.5593326 = fieldWeight in 2210, product of:
               2.4494898 = tf(freq=6.0), with freq of:
                 6.0 = termFreq=6.0
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.0625 = fieldNorm(doc=2210)
         0.04034447 = weight(abstract_txt:novel in 2210) [ClassicSimilarity], result of:
           0.04034447 = score(doc=2210,freq=1.0), product of:
             0.116742305 = queryWeight, product of:
               1.3220754 = boost
               5.529371 = idf(docFreq=476, maxDocs=44218)
               0.015969684 = queryNorm
             0.34558567 = fieldWeight in 2210, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.529371 = idf(docFreq=476, maxDocs=44218)
               0.0625 = fieldNorm(doc=2210)
         0.070912294 = weight(abstract_txt:respectively in 2210) [ClassicSimilarity], result of:
           0.070912294 = score(doc=2210,freq=1.0), product of:
             0.1700278 = queryWeight, product of:
               1.5955193 = boost
               6.6730065 = idf(docFreq=151, maxDocs=44218)
               0.015969684 = queryNorm
             0.4170629 = fieldWeight in 2210, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               6.6730065 = idf(docFreq=151, maxDocs=44218)
               0.0625 = fieldNorm(doc=2210)
         0.033739667 = weight(abstract_txt:framework in 2210) [ClassicSimilarity], result of:
           0.033739667 = score(doc=2210,freq=1.0), product of:
             0.118621446 = queryWeight, product of:
               1.6321847 = boost
               4.550903 = idf(docFreq=1268, maxDocs=44218)
               0.015969684 = queryNorm
             0.28443143 = fieldWeight in 2210, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.550903 = idf(docFreq=1268, maxDocs=44218)
               0.0625 = fieldNorm(doc=2210)
         0.72732764 = weight(abstract_txt:sentiment in 2210) [ClassicSimilarity], result of:
           0.72732764 = score(doc=2210,freq=8.0), product of:
             0.54468113 = queryWeight, product of:
               4.515265 = boost
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.015969684 = queryNorm
             1.3353274 = fieldWeight in 2210, product of:
               2.828427 = tf(freq=8.0), with freq of:
                 8.0 = termFreq=8.0
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.0625 = fieldNorm(doc=2210)
       0.24 = coord(6/25)
   

2. Wei, W.; Liu, Y.-P.; Wei, L-R.: Feature-level sentiment analysis based on rules and fine-grained domain ontology (2020) 0.20

   0.1963889 = sum of:
     0.1963889 = product of:
       0.9819445 = sum of:
         0.024385227 = weight(abstract_txt:context in 5876) [ClassicSimilarity], result of:
           0.024385227 = score(doc=5876,freq=1.0), product of:
             0.071920075 = queryWeight, product of:
               1.0376889 = boost
               4.339969 = idf(docFreq=1566, maxDocs=44218)
               0.015969684 = queryNorm
             0.3390601 = fieldWeight in 5876, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.339969 = idf(docFreq=1566, maxDocs=44218)
               0.078125 = fieldNorm(doc=5876)
         0.027146412 = weight(abstract_txt:level in 5876) [ClassicSimilarity], result of:
           0.027146412 = score(doc=5876,freq=1.0), product of:
             0.07725155 = queryWeight, product of:
               1.0754635 = boost
               4.497956 = idf(docFreq=1337, maxDocs=44218)
               0.015969684 = queryNorm
             0.3514028 = fieldWeight in 5876, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.497956 = idf(docFreq=1337, maxDocs=44218)
               0.078125 = fieldNorm(doc=5876)
         0.037797417 = weight(abstract_txt:analysis in 5876) [ClassicSimilarity], result of:
           0.037797417 = score(doc=5876,freq=3.0), product of:
             0.07645344 = queryWeight, product of:
               1.3103467 = boost
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.015969684 = queryNorm
             0.4943848 = fieldWeight in 5876, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.078125 = fieldNorm(doc=5876)
         0.042174585 = weight(abstract_txt:framework in 5876) [ClassicSimilarity], result of:
           0.042174585 = score(doc=5876,freq=1.0), product of:
             0.118621446 = queryWeight, product of:
               1.6321847 = boost
               4.550903 = idf(docFreq=1268, maxDocs=44218)
               0.015969684 = queryNorm
             0.3555393 = fieldWeight in 5876, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.550903 = idf(docFreq=1268, maxDocs=44218)
               0.078125 = fieldNorm(doc=5876)
         0.85044086 = weight(abstract_txt:sentiment in 5876) [ClassicSimilarity], result of:
           0.85044086 = score(doc=5876,freq=7.0), product of:
             0.54468113 = queryWeight, product of:
               4.515265 = boost
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.015969684 = queryNorm
             1.5613555 = fieldWeight in 5876, product of:
               2.6457512 = tf(freq=7.0), with freq of:
                 7.0 = termFreq=7.0
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.078125 = fieldNorm(doc=5876)
       0.2 = coord(5/25)
   

3. Nguyen, T.T.; Tho Thanh Quan, T.T.; Tuoi Thi Phan, T.T.: Sentiment search : an emerging trend on social media monitoring systems (2014) 0.19

   0.19066267 = sum of:
     0.19066267 = product of:
       0.95331335 = sum of:
         0.019508181 = weight(abstract_txt:context in 1625) [ClassicSimilarity], result of:
           0.019508181 = score(doc=1625,freq=1.0), product of:
             0.071920075 = queryWeight, product of:
               1.0376889 = boost
               4.339969 = idf(docFreq=1566, maxDocs=44218)
               0.015969684 = queryNorm
             0.27124807 = fieldWeight in 1625, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.339969 = idf(docFreq=1566, maxDocs=44218)
               0.0625 = fieldNorm(doc=1625)
         0.024689171 = weight(abstract_txt:analysis in 1625) [ClassicSimilarity], result of:
           0.024689171 = score(doc=1625,freq=2.0), product of:
             0.07645344 = queryWeight, product of:
               1.3103467 = boost
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.015969684 = queryNorm
             0.3229308 = fieldWeight in 1625, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.0625 = fieldNorm(doc=1625)
         0.058438815 = weight(abstract_txt:framework in 1625) [ClassicSimilarity], result of:
           0.058438815 = score(doc=1625,freq=3.0), product of:
             0.118621446 = queryWeight, product of:
               1.6321847 = boost
               4.550903 = idf(docFreq=1268, maxDocs=44218)
               0.015969684 = queryNorm
             0.49264967 = fieldWeight in 1625, product of:
               1.7320508 = tf(freq=3.0), with freq of:
                 3.0 = termFreq=3.0
               4.550903 = idf(docFreq=1268, maxDocs=44218)
               0.0625 = fieldNorm(doc=1625)
         0.12334952 = weight(abstract_txt:aspect in 1625) [ClassicSimilarity], result of:
           0.12334952 = score(doc=1625,freq=2.0), product of:
             0.22343488 = queryWeight, product of:
               2.2400782 = boost
               6.2458487 = idf(docFreq=232, maxDocs=44218)
               0.015969684 = queryNorm
             0.55206025 = fieldWeight in 1625, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               6.2458487 = idf(docFreq=232, maxDocs=44218)
               0.0625 = fieldNorm(doc=1625)
         0.72732764 = weight(abstract_txt:sentiment in 1625) [ClassicSimilarity], result of:
           0.72732764 = score(doc=1625,freq=8.0), product of:
             0.54468113 = queryWeight, product of:
               4.515265 = boost
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.015969684 = queryNorm
             1.3353274 = fieldWeight in 1625, product of:
               2.828427 = tf(freq=8.0), with freq of:
                 8.0 = termFreq=8.0
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.0625 = fieldNorm(doc=1625)
       0.2 = coord(5/25)
   

4. Xu, L.; Qiu, J.: Unsupervised multi-class sentiment classification approach (2019) 0.18

   0.18323378 = sum of:
     0.18323378 = product of:
       0.91616887 = sum of:
         0.01745788 = weight(abstract_txt:analysis in 5003) [ClassicSimilarity], result of:
           0.01745788 = score(doc=5003,freq=1.0), product of:
             0.07645344 = queryWeight, product of:
               1.3103467 = boost
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.015969684 = queryNorm
             0.22834657 = fieldWeight in 5003, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.0625 = fieldNorm(doc=5003)
         0.04034447 = weight(abstract_txt:novel in 5003) [ClassicSimilarity], result of:
           0.04034447 = score(doc=5003,freq=1.0), product of:
             0.116742305 = queryWeight, product of:
               1.3220754 = boost
               5.529371 = idf(docFreq=476, maxDocs=44218)
               0.015969684 = queryNorm
             0.34558567 = fieldWeight in 5003, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.529371 = idf(docFreq=476, maxDocs=44218)
               0.0625 = fieldNorm(doc=5003)
         0.050779503 = weight(abstract_txt:accuracy in 5003) [ClassicSimilarity], result of:
           0.050779503 = score(doc=5003,freq=1.0), product of:
             0.13609158 = queryWeight, product of:
               1.4274395 = boost
               5.9700394 = idf(docFreq=306, maxDocs=44218)
               0.015969684 = queryNorm
             0.37312746 = fieldWeight in 5003, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               5.9700394 = idf(docFreq=306, maxDocs=44218)
               0.0625 = fieldNorm(doc=5003)
         0.036139604 = weight(abstract_txt:time in 5003) [ClassicSimilarity], result of:
           0.036139604 = score(doc=5003,freq=2.0), product of:
             0.09856323 = queryWeight, product of:
               1.4878021 = boost
               4.148331 = idf(docFreq=1897, maxDocs=44218)
               0.015969684 = queryNorm
             0.36666414 = fieldWeight in 5003, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               4.148331 = idf(docFreq=1897, maxDocs=44218)
               0.0625 = fieldNorm(doc=5003)
         0.7714474 = weight(abstract_txt:sentiment in 5003) [ClassicSimilarity], result of:
           0.7714474 = score(doc=5003,freq=9.0), product of:
             0.54468113 = queryWeight, product of:
               4.515265 = boost
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.015969684 = queryNorm
             1.4163285 = fieldWeight in 5003, product of:
               3.0 = tf(freq=9.0), with freq of:
                 9.0 = termFreq=9.0
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.0625 = fieldNorm(doc=5003)
       0.2 = coord(5/25)
   

5. Saif, H.; He, Y.; Fernandez, M.; Alani, H.: Contextual semantics for sentiment analysis of Twitter (2016) 0.17

   0.16795948 = sum of:
     0.16795948 = product of:
       0.8397974 = sum of:
         0.019508181 = weight(abstract_txt:context in 2667) [ClassicSimilarity], result of:
           0.019508181 = score(doc=2667,freq=1.0), product of:
             0.071920075 = queryWeight, product of:
               1.0376889 = boost
               4.339969 = idf(docFreq=1566, maxDocs=44218)
               0.015969684 = queryNorm
             0.27124807 = fieldWeight in 2667, product of:
               1.0 = tf(freq=1.0), with freq of:
                 1.0 = termFreq=1.0
               4.339969 = idf(docFreq=1566, maxDocs=44218)
               0.0625 = fieldNorm(doc=2667)
         0.04343426 = weight(abstract_txt:level in 2667) [ClassicSimilarity], result of:
           0.04343426 = score(doc=2667,freq=4.0), product of:
             0.07725155 = queryWeight, product of:
               1.0754635 = boost
               4.497956 = idf(docFreq=1337, maxDocs=44218)
               0.015969684 = queryNorm
             0.5622445 = fieldWeight in 2667, product of:
               2.0 = tf(freq=4.0), with freq of:
                 4.0 = termFreq=4.0
               4.497956 = idf(docFreq=1337, maxDocs=44218)
               0.0625 = fieldNorm(doc=2667)
         0.024689171 = weight(abstract_txt:analysis in 2667) [ClassicSimilarity], result of:
           0.024689171 = score(doc=2667,freq=2.0), product of:
             0.07645344 = queryWeight, product of:
               1.3103467 = boost
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.015969684 = queryNorm
             0.3229308 = fieldWeight in 2667, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               3.6535451 = idf(docFreq=3112, maxDocs=44218)
               0.0625 = fieldNorm(doc=2667)
         0.07181306 = weight(abstract_txt:accuracy in 2667) [ClassicSimilarity], result of:
           0.07181306 = score(doc=2667,freq=2.0), product of:
             0.13609158 = queryWeight, product of:
               1.4274395 = boost
               5.9700394 = idf(docFreq=306, maxDocs=44218)
               0.015969684 = queryNorm
             0.5276819 = fieldWeight in 2667, product of:
               1.4142135 = tf(freq=2.0), with freq of:
                 2.0 = termFreq=2.0
               5.9700394 = idf(docFreq=306, maxDocs=44218)
               0.0625 = fieldNorm(doc=2667)
         0.6803527 = weight(abstract_txt:sentiment in 2667) [ClassicSimilarity], result of:
           0.6803527 = score(doc=2667,freq=7.0), product of:
             0.54468113 = queryWeight, product of:
               4.515265 = boost
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.015969684 = queryNorm
             1.2490844 = fieldWeight in 2667, product of:
               2.6457512 = tf(freq=7.0), with freq of:
                 7.0 = termFreq=7.0
               7.5537524 = idf(docFreq=62, maxDocs=44218)
               0.0625 = fieldNorm(doc=2667)
       0.2 = coord(5/25)