Zipf’s Analysis among Different Languages--论文代写范文精选

有趣的是,最近的研究已经表明我们Zipf定律不仅出现在文本对象的统计分析。例如,Zipf定律也发现在DNA序列,城市人口，地球物理学等,甚至激发了另一种投资策略分析。下面的essay代写范文将进行详述。

Abstract
The paper introduces the used of Zipfian statistics to observe the human languages by using the same (meaning) corpus/corpora but different in grammatical and structural utterances. We used biblical texts since they contain corpuses that have been most widely and carefully translated into many languages. The idea is to reduce the possibility of noise came from the   meaning of the texts in distinctive language. The result is that the robustness of the Zipfian law is observable and some statistical differences are discovered between English and widely used national and several ethnic languages in Indonesia. The paper ends by modestly propose further possible framework in interdisciplinary approaches to human language evolution. Keywords: statistical processing of natural language, Zipf’s law, Zipf‐Mandelbrot fit, corpus, evolution of language.

Introduction
There have been a lot of places beyond many scientific domains exhibited the signatures of power law as an interesting statistical properties emerged from complex systems (SItungkir & Surya, 2003). Concerning the classic work of Harvard linguistic professor, G. K. Zipf (1947), the power law seems to be not that outlandish among linguist, as it has been showed that the power law in the Zipf plot is linked to the writer and reader natural behavior to minimizing their effort while communicating.

In this case, the writer’s Page 2 of 11 efforst is conserved by using a small vocabulary of common words and the reader’s effort is reduced by having a large vocabulary of individually rarer words – a way to make the messages less vague (Manning, et. al., 1999). Here, Zipf argued that the Zipf’s law is supported by the maximally economical compromise between the competing needs between writer (or speaker) and reader (or listener).   Interestingly, recent studies on astonishingly different fields of research have shown us that the Zipf’s Law is not only emerged in statistical analysis of textual objects. For instance, the Zipf’s law is also discovered in the DNA sequences (Mantegna, et. al., 1994), population of cities (Mulianta, et. al., 2004), rank of general elections (Situngkir & Surya, 2004), the daily precipitation series in geophysics (Primo, et. al., 2007) and even inspired an alternative to investment strategies (Situngkir & Surya, 2005). Whether or not the persistence of the law in those fields related to its origin in linguistics are basically left to philosophical issues may arise but yet, we can now see that the (statistical) application of the Zipf’s law are not limited to a single discipline of study.

Discussions
Zipf Plot over Languages Talking about the bible, it is in fact coalesced by 66 books, 39 known as the Old Testament and 27 as the New Testament. Those books are considered written by different persons except, for example, the first five books of the Old Testament, literally known as the Pentateuch or the majority books in the new testament recognized written by St. Paul. Regarding to the sizes of the books and understanding the nature of Zipf’s law as “macro” properties with some requirements of (statistically) large enough data, we realize that it is obvious that Zipf’s law is deviated for some short books compiled in the bible. We do the observations of the Zipf and Zipf‐Mandelbrot fit and analysis; however, some of them are displayed in the paper to ease us to see the emerging patterns.

As an exemplification, we display the Zipf’s plot in figure 2 and it showed that all of the texts exhibit the power‐law with the exponent close to unity. However, from the figure 1, we can also see that there are some slight differences among them – especially when we compare the English version with the Indonesian and Indonesian ethnic languages. This interesting feature can be observed even more clearly in figure 3 (and of course quantitatively in table 1). Throughout our observation, the data of national Indonesian language will “slightly deviated” more to the English language while it will just “close enough” to the ethnic ones. This interesting discussion surely could bring us to the conjecture that there is some different statistical properties that can be differed among multiple language corpus.

Concluding Remarks
The paper reports the statistical observation of Zipf’s law to different human languages while the approached corpus is being telling the same things. This is expected to reduce the possible sensitivity to the meaning of the texts and the different stylized statistics are closer to what emerging from the respective structure of language, whether it grammatical or lexical. Interestingly, it has also been showed that Zipfian statistics is robust throughout those raw corpuses we analyzed.    From the comparative analysis, we point out that there are some possible conjectures of statistical distinctions between languages and this may open a good challenge in our understanding to the broader sense of our scientific recognition to nature of human language. Since different languages are sensitive to the evolution of human species with very tight relationship with cultural phenomena, social environments, and even geographical and natural system in which the language widely used and developed, this can bring us further to the evolution of language and optimistically reveal its evolutionary life history. We modestly point out here that human utterances can be observed by its quantitative nature by using better and better statistical tools. （essay代写)

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。（essay代写)
更多essay代写范文欢迎访问我们主页 www.51due.com 当然有essay代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X（essay代写)