Dresden 2017 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 21: Social Systems II
SOE 21.6: Talk
Thursday, March 23, 2017, 18:15–18:30, GÖR 226
Innovation- and information production rate for sentences of particular length — •Bo Liu1, Stefan Thurner1,2,3,4, Rudolf Hanel1, and Bernat Corominas-Murtra1 — 1Section for Science of Complex Systems, Medical University of Vienna, Spitalgasse 23, A-1090, Austria — 2Santa Fe Institute, 1399 Hyde Park Road, Santa Fe, NM 87501, USA — 3International Institute for Applied Systems Analysis, Schlossplatz 1, A-2361 Laxenburg, Austria — 4Complexity Science Hub Vienna, Josefstädter Straße 39, A-1080, Austria
Innovations are part of our lives and are the engines that boost our society. The understanding of the underlying dynamics is therefore essential. Language has been considered as a relatively simple toy model to study innovation dynamics. Information in language is encoded in units of different sizes: letters, words, sentences and paragraph. While at the level of letters, many results on information production rate exist, on the level of sentences much less is known. A simple measure of ``innovation rate'' in language is the so-called Heaps' exponent. We investigate subtexts which are composed of sentences with a particular length (number of words). A non-monotonic behavior of the Heaps' exponent vs. sentence lengths is found, with a maximum value at around sentence length 7. Similar behavior appears in the Zipf exponent and the cross entropy, which measures the information production rate. We analyze texts of the Corpus of Historical American English (CoHA) from 1800 to 2000 and find that the discovered pattern is slightly becoming stronger across history.