arXivDaily arXiv每日学术速递 周一至周五更新
arXiv physics/0511049physics.soc-ph
收藏

物理学家模拟语言竞赛

Simulation of language competition by physicists

Christian Schulze, Dietrich Stauffer

摘要:

在Abrams和Strogatz 2003以及Patriarca和Leppanen 2004之后,其他五个物理小组独立地开始模拟语言的竞争,而不是从猿的声音中进化出人类语言,或者儿童学习语言。这次演讲集中在Christian Schulze等人和Viviane de Oliveira等人的模型上,这些模型允许模拟大量语言,类似于今天的8,000种人类语言。第一个模型处理的是一种语言的随机突变,从其他语言迁移到其他语言,以及逃离只有少数人使用的语言的连续过程。第二个模型结合了这些飞行和突变方面,忽略了转移,并描述了一个大的地理区域的殖民地的人开始在一个格子点。一种语言的规模是由说这种语言的人数来定义的,第一个模型给出了一个真实的语言规模的对数正态分布,但是这个数字是不好的。对于第二个模型,我们的蒙特卡罗模拟给出了高达10亿的大小,但不是接近对数正态的形状。计划于2006年9月中旬在波兰举行一次会议。

英文摘要:

Following Abrams and Strogatz 2003 and Patriarca and Leppanen 2004, five other physics groups independently started to simulate the competition of languages, as opposed to the evolution of a human language out of ape sounds, or the learning of a language by a child. This talk concentrates on the models of Christian Schulze et al and of Viviane de Oliveira et al which allow the simulation of a large number of languages, similar to today's 8,000 human languages. The first model deals with a continuous process of random mutations of a language, transfer from and to other languages, and flight away from languages spoken by only a few people. The second model combines these flight and mutation aspects, ignores transfer and describes the colonization of a large geographical region by people starting at one lattice point. The size of a language is defined by the number of people speaking it. The first model gives a realistic log-normal shape for the histogram of language sizes but the numbers are bad. For the second model our Monte Carlo simulations give sizes up to thousand million, but not the nearly log-normal shape. A meeting is planned for mid-September 2006 in Poland.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑