Journal of the Association for Information Science



Bert R. Boyce


Special Topic Issue: Knowledge Discovery and Data Mining
Guest Editors:
Vijay V. Raghavan, Jitender S. Deogun, and Hayri Sever


Vijay V. Raghavan, Jitender S. Deogun, and Hayri Sever




Computational Methods for Rough Classification and Discovery
D. A. Bell and J. W. Guan




Data Mining Using Extensions of the Rough Set Model
P. J. Lingras and Y. Y. Yao




Feature Selection and Effective Classifiers
Jitender S. Deogun, Suresh K. Choubey, Vijay V. Raghavan, and Hayri Sever




Rule Induction with Extension Matrices
Xindong Wu




Automated Database Schema Design Using Mined Data Dependencies
S. K. M. Wong, C. J. Butz, and Y. Xiang




Non-Indexed Indirect-Collective Citedness (NIICC)
Endre Száva-Kováts

By "indirect-collective" references, Száva-Kováts means such instances as those in which an author refers to, "the references contained therein," when referring to another source. This brief communication describes a review of 331 Physical Review articles, and 290 Journal of the Optical Society of America articles from 1969 which found 49 such instances in Physical Review (PHR) and 21 in Journal of the Optical Society of America (JOSA). If these articles are checked for "indirect-collective" references of a secondary nature the set will expand. One article chosen for multi-level iterative examination had 18 formal references, 220 indirect-collectively cited references, and 8 additional eponymal references. Another with 3 formal references produced 124 indirect-collectively cited references. This phenomenon prejudices the completeness of formal citation at least during the early Big Science period in physics.




Cumulative Advantage and Success-Breeds-Success: The Value of Time Pattern Analysis
John C. Huber

Huber finds that observable bibliometric data fit not only the Cumulative Advantage distribution, but also the Allison model, here called the non-uniform giftedness distribution, which considers the publication rate of an individual author to be static, and the distribution of publications over authors to be determined by the rate of publication over the population. Thus Cumulative Advantage can be tested by determining the time pattern of publication of individual authors. Looking at a random sample of inventors, the distribution of individual patent publication fits a Poisson distribution whose parameter is constant over the time interval, which rejects the Cumulative Advantage distribution hypothesis for this data unless some debilitating process is also in effect


ASIS HomeSearch ASISMake A Comment

© 1998 , Association for Information Science
Last update: November 06, 1998