Sentence-Length Distributions of Greek Authors

Abstract

9.—Wealth and Taxable Capacity of India. By K. T. Shah, Professor of Economics, University of Bombay, and K. J. Kham‐bata, M.A. xxi + 347 pp. Bombay: D. P. Taraporevala; London: P. S. King, 1924. Price 15s. net. The Wealth of India. By P. A. Wadia, Professor of Politics and History, Wilson College, Bombay, and G. N. Joshi, Lecturer in Economics, Wilson College, Bombay. xi +438 pp. Macmillan, 1925. Price 21s. net.

Adaptation of fictional and online conversations to communication media

“…More recently, sentence length has been used to classify text genre by itself [17] or in combination with other text properties [18]. Yule's 1939 paper did not provide the sentence length distribution but several decades after its publication, the distribution was described as lognormal [19,20,21] which was later shown by Sichel [22,23] to be flawed. More recently, Sigurd et al [24] showed that sentence length distributions may be approximated by a gamma distribution.…”

Authorship attribution

“…Williams (1940) discovered that by taking a frequency distribution of the logarithm of the number of words per sentence, an approximation to a Normal distribution was found for each author. The lognormal model suggested by Williams and used by Wake (1957) must be rejected on several grounds, one of which is that most of the observed sentence-length distributions, after logarithmic transformation, are negatively skew. Sichel (1974) suggests a compound Poisson distribution for representing sentence-length distributions which appears to work well.…”

On a Distribution Representing Sentence-Length in Written Prose

“…Some sentence-length distributions do approach a straight line on log-probability paper. To check whether such cases can be represented satisfactorily by distribution (8), two frequency counts given by Wake (1957) were fitted to (8) and they are shown in Table 4. The first example refers to sentencelengths from Timaeus by Plato and the second comes from the Hippocratic Corpus, Regimen in Acute Diseases.…”

“…The lognormal model suggested by Williams and used by Wake must be rejected on several grounds: In the first place the number of words in a sentence constitutes a discrete variable whereas the lognormal distribution is continuous. Wake (1957) has pointed out that most observed log-sentence-Iength distributions display upper tails which tend towards zero much faster than the corresponding normal distribution. This is also evident in most of the cumulative percentage frequency distributions of sentence-lengths plotted on log-probability paper by Williams (1970).…”