Search (2 results, page 1 of 1)

  • × year_i:[2000 TO 2010}
  • × author_ss:"Watters, C."
  1. Watters, C.; Wang, H.: Rating new documents for similarity (2000) 0.06 
    0.056825537 = product of:
  0.113651074 = sum of:
    0.113651074 = product of:
      0.22730215 = sum of:
        0.22730215 = weight(_text_:news in 4856) [ClassicSimilarity], result of:
          0.22730215 = score(doc=4856,freq=12.0), product of:
            0.26705483 = queryWeight, product of:
              5.2416887 = idf(docFreq=635, maxDocs=44218)
              0.05094824 = queryNorm
            0.85114413 = fieldWeight in 4856, product of:
              3.4641016 = tf(freq=12.0), with freq of:
                12.0 = termFreq=12.0
              5.2416887 = idf(docFreq=635, maxDocs=44218)
              0.046875 = fieldNorm(doc=4856)
      0.5 = coord(1/2)
  0.5 = coord(1/2)
    Abstract
    Electronic news has long held the promise of personalized and dynamic delivery of current event new items, particularly for Web users. Although wlwctronic versions of print news are now widely available, the personalization of that delivery has not yet been accomplished. In this paper, we present a methodology of associating news documents based on the extraction of feature phrases, where feature phrases identify dates, locations, people and organizations. A news representation is created from these feature phrases to define news objects that can then be compared and ranked to find related news items. Unlike tradtional information retrieval, we are much more interested in precision than recall. That is, the user would like to see one or more specifically related articles, rather than all somewhat related articles. The algorithm is designed to work interactively the the user using regular web browsers as the interface
  2. Shepherd, M.; Duffy, J.F.J.; Watters, C.; Gugle, N.: ¬The role of user profiles for news filtering (2001) 0.05 
    0.05114883 = product of:
  0.10229766 = sum of:
    0.10229766 = product of:
      0.20459533 = sum of:
        0.20459533 = weight(_text_:news in 5585) [ClassicSimilarity], result of:
          0.20459533 = score(doc=5585,freq=14.0), product of:
            0.26705483 = queryWeight, product of:
              5.2416887 = idf(docFreq=635, maxDocs=44218)
              0.05094824 = queryNorm
            0.76611733 = fieldWeight in 5585, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              5.2416887 = idf(docFreq=635, maxDocs=44218)
              0.0390625 = fieldNorm(doc=5585)
      0.5 = coord(1/2)
  0.5 = coord(1/2)
    Abstract
    Most on-line news sources are electronic versions of "ink-on-paper" newspapers. These are versions that have been filtered, from the mass of news produced each day, by an editorial board with a given community profile in mind. As readers, we choose the filter rather than choose the stories. New technology, however, provides the potential for personalized versions to be filtered automatically from this mass of news on the basis of user profiles. People read the news for many reasons: to find out "what's going on," to be knowledgeable members of a community, and because the activity itself is pleasurable. Given this, we ask the question, "How much filtering is acceptable to readers?" In this study, an evaluation of user preference for personal editions versus community editions of on-line news was performed. A personalized edition of a local newspaper was created for each subject based on an elliptical model that combined the user profile and community profile as represented by the full edition of the local newspaper. The amount of emphasis given the user profile and the community profile was varied to test the subjects' reactions to different amounts of personalized filtering. The task was simply, "read the news," rather than any subject specific information retrieval task. The results indicate that users prefer the coarse-grained community filters to fine-grained personalized filters

Authors