Search (3 results, page 1 of 1)

  • × author_ss:"An, L."
  1. Yu, C.; Xue, H.; An, L.; Li, G.: ¬A lightweight semantic-enhanced interactive network for efficient short-text matching (2023) 0.02 
    0.018982807 = product of:
  0.037965614 = sum of:
    0.037965614 = sum of:
      0.006765375 = weight(_text_:a in 890) [ClassicSimilarity], result of:
        0.006765375 = score(doc=890,freq=8.0), product of:
          0.053105544 = queryWeight, product of:
            1.153047 = idf(docFreq=37942, maxDocs=44218)
            0.046056706 = queryNorm
          0.12739488 = fieldWeight in 890, product of:
            2.828427 = tf(freq=8.0), with freq of:
              8.0 = termFreq=8.0
            1.153047 = idf(docFreq=37942, maxDocs=44218)
            0.0390625 = fieldNorm(doc=890)
      0.03120024 = weight(_text_:22 in 890) [ClassicSimilarity], result of:
        0.03120024 = score(doc=890,freq=2.0), product of:
          0.16128273 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.046056706 = queryNorm
          0.19345059 = fieldWeight in 890, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.0390625 = fieldNorm(doc=890)
  0.5 = coord(1/2)
    Abstract
    Knowledge-enhanced short-text matching has been a significant task attracting much attention in recent years. However, the existing approaches cannot effectively balance effect and efficiency. Effective models usually consist of complex network structures leading to slow inference speed and the difficulties of applications in actual practice. In addition, most knowledge-enhanced models try to link the mentions in the text to the entities of the knowledge graphs-the difficulties of entity linking decrease the generalizability among different datasets. To address these problems, we propose a lightweight Semantic-Enhanced Interactive Network (SEIN) model for efficient short-text matching. Unlike most current research, SEIN employs an unsupervised method to select WordNet's most appropriate paraphrase description as the external semantic knowledge. It focuses on integrating semantic information and interactive information of text while simplifying the structure of other modules. We conduct intensive experiments on four real-world datasets, that is, Quora, Twitter-URL, SciTail, and SICK-E. Compared with state-of-the-art methods, SEIN achieves the best performance on most datasets. The experimental results proved that introducing external knowledge could effectively improve the performance of the short-text matching models. The research sheds light on the role of lightweight models in leveraging external knowledge to improve the effect of short-text matching.
    Date
    22. 1.2023 19:05:27
    Type
    a
  2. Zhang, J.; An, L.; Tang, T.; Hong, Y.: Visual health subject directory analysis based on users' traversal activities (2009) 0.00 
    0.0024857575 = product of:
  0.004971515 = sum of:
    0.004971515 = product of:
      0.00994303 = sum of:
        0.00994303 = weight(_text_:a in 3112) [ClassicSimilarity], result of:
          0.00994303 = score(doc=3112,freq=12.0), product of:
            0.053105544 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046056706 = queryNorm
            0.18723148 = fieldWeight in 3112, product of:
              3.4641016 = tf(freq=12.0), with freq of:
                12.0 = termFreq=12.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=3112)
      0.5 = coord(1/2)
  0.5 = coord(1/2)
    Abstract
    Concerns about health issues cover a wide spectrum. Consumer health information, which has become more available on the Internet, plays an extremely important role in addressing these concerns. A subject directory as an information organization and browsing mechanism is widely used in consumer health-related Websites. In this study we employed the information visualization technique Self-Organizing Map (SOM) in combination with a new U-matrix algorithm to analyze health subject clusters through a Web transaction log. An experimental study was conducted to test the proposed methods. The findings show that the clusters identified from the same cells based on path-length-1 outperformed both the clusters from the adjacent cells based on path-length-1 and the clusters from the same cells based on path-length-2 in the visual SOM display. The U-matrix method successfully distinguished the irrelevant subjects situated in the adjacent cells with different colors in the SOM display. The findings of this study lead to a better understanding of the health-related subject relationship from the users' traversal perspective.
    Type
    a
  3. An, L.; Zhang, J.; Yu, C.: ¬The visual subject analysis of library and information science journals with self-organizing map (2011) 0.00 
    0.0016913437 = product of:
  0.0033826875 = sum of:
    0.0033826875 = product of:
      0.006765375 = sum of:
        0.006765375 = weight(_text_:a in 4613) [ClassicSimilarity], result of:
          0.006765375 = score(doc=4613,freq=8.0), product of:
            0.053105544 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046056706 = queryNorm
            0.12739488 = fieldWeight in 4613, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=4613)
      0.5 = coord(1/2)
  0.5 = coord(1/2)
    Abstract
    Academic journals play an important role in scientific communication. The effective organization of journals can help reveal the thematic contents of journals and thus make them more user-friendly. In this study, the Self-Organizing Map (SOM) technique was employed to visually analyze the 60 library and information science-related journals published from 2006 to 2008. The U-matrix by Ultsch (2003) was applied to categorize the journals into 19 clusters according to their subjects. Four journals were recommended to supplement library collections although they were not indexed by SCI/SSCI. A novel SOM display named Attribute Accumulation Matrix (AA-matrix) was proposed, and the results from this method show that they correlate significantly with the total occurrences of the subjects in the investigated journals. The AA-matrix was employed to identify the 86 salient subjects, which could be manually classified into 7 meaningful groups. A method of the Salient Attribute Projection was constructed to label the attribute characteristics of different clusters. Finally, the subject characteristics of the journals with high impact factors (IFs) were also addressed. The findings of this study can lead to a better understanding of the subject structure and characteristics of library/information-related journals.
    Type
    a

Authors

Years