| I.H. Witten, T.C. Bell, A. Mo#at, C.G. NevillManning, T.C. Smith, and H. Thimbleby. Semantic and generative models for lossy text compression. Computer Journal, 37(2):83--87, 1994. |
....C, G, and T) occur with frequency greater than 1 ; these can be represented with 2 bit codes. The other eleven, wildcard characters are stored as #character, position list# pairs, indicating the positions at which each wildcard The resemblance between this technique and lossy text compression [11] is purely superficial. pair models. Word model 28.4 17.97 1.20 29.9 258.48 5.26 All pairs 37.0 23.42 17.37 32.9 284.21 110.90 Pair model, P = 10 27.1 17.14 1.88 28.6 246.86 12.26 Pair model, P = 1000 27.8 17.59 1.20 28.8 248.66 5.33 can be substituted. A randomly chosen nucleotide is ....
I.H. Witten, T.C. Bell, A. Mo#at, C.G. NevillManning, T.C. Smith, and H. Thimbleby. Semantic and generative models for lossy text compression. Computer Journal, 37(2):83--87, 1994.
.... Lossy compression allows to obtain significantly higher compression ratios preserving a representative subset of original data; interesting approaches to lossy compression are wavelet transformations [11] histograms [9, 12] and methods for the extraction of significant parts from a free text [13]. XML uses markups to identify and describe data; the schema related information is contained in documents themselves so the language is called self describing. Therefore, it is always possible to directly identify data through their paths; this feature becomes specially useful when when data ....
Witten, I.H., et al., "Semantic and generative models for lossy text compression", The Computer Journal, Volume 37, Issue 2, 1994.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC