Search (2 results, page 1 of 1)

  • × author_ss:"Kuflik, T."
  • × language_ss:"e"
  • × year_i:[2000 TO 2010}
  1. Kuflik, T.; Shapira, B.; Shoval, P.: Stereotype-based versus personal-based filtering rules in information filtering systems (2003) 0.03 
    0.029261088 = product of:
  0.14630544 = sum of:
    0.14630544 = weight(_text_:mail in 1234) [ClassicSimilarity], result of:
      0.14630544 = score(doc=1234,freq=4.0), product of:
        0.28137597 = queryWeight, product of:
          5.5462847 = idf(docFreq=468, maxDocs=44218)
          0.050732337 = queryNorm
        0.5199642 = fieldWeight in 1234, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.5462847 = idf(docFreq=468, maxDocs=44218)
          0.046875 = fieldNorm(doc=1234)
  0.2 = coord(1/5)
    Abstract
    Kuflick, et alia, test whether an e-mail filter based on personally designed rules will be as effective as one whose rules are designed to reflect the average user in a specified group of users. Using a prototype filtering system ten subjects were interviewed to construct their own personal rules and were also assigned to one of four predefined rule sets generated by cluster analysis from 40 interviews using the same instrument with like subjects. Assignment was based upon social parameters such as education, profession, and computer knowledge level in the data gathered. The rules led to assignment of a relevance number in the range 1 to 7 to each message based upon the participant chosen values of goal, length, and history parameters of the message. A set of e-mail messages was then supplied to the 10 subjects who ranked them as to relevance. Pearson coefficients between personal rule ranks and user ranks are consistently lower than the correlations between user ranks and the stereotype ranks but in only three cases significantly so.
  2. Goren-Bar, D.; Kuflik, T.: Supporting user-subjective categorization with self-organizing maps and learning vector quantization (2005) 0.02 
    0.01724226 = product of:
  0.0862113 = sum of:
    0.0862113 = weight(_text_:mail in 3325) [ClassicSimilarity], result of:
      0.0862113 = score(doc=3325,freq=2.0), product of:
        0.28137597 = queryWeight, product of:
          5.5462847 = idf(docFreq=468, maxDocs=44218)
          0.050732337 = queryNorm
        0.30639184 = fieldWeight in 3325, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.5462847 = idf(docFreq=468, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3325)
  0.2 = coord(1/5)
    Abstract
    Today, most document categorization in organizations is done manually. We save at work hundreds of files and e-mail messages in folders every day. While automatic document categorization has been widely studied, much challenging research still remains to support usersubjective categorization. This study evaluates and compares the application of self-organizing maps (SOMs) and learning vector quantization (LVO) with automatic document classification, using a set of documents from an organization, in a specific domain, manually classified by a domain expert. After running the SOM and LVO we requested the user to reclassify documents that were misclassified by the system. Results show that despite the subjective nature of human categorization, automatic document categorization methods correlate weIl with subjective, personal categorization, and the LVO method outperforms the SOM. The reclassification process revealed an interesting pattern: About 40% of the documents were classified according to their original categorization, about 35% according to the system's categorization (the users changed the original categorization), and the remainder received a different (new) categorization. Based an these results we conclude that automatic support for subjective categorization is feasible; however, an exact match is probably impossible due to the users' changing categorization behavior.

Authors