【摘要】：以26個(gè)漢語方言點(diǎn)古全濁聲母的今讀為樣品,用NTSYS進(jìn)行聚類分析和主坐標(biāo)分析,并討論相關(guān)的問題。結(jié)論是:(1)聚類分析結(jié)果不一定能直接作為一種合理的分類方案;(2)古全濁聲母演變的特定方式跟大方言之間不存在一對(duì)一的關(guān)系;(3)漢語方言基本特征庫建設(shè)的困難主要在于轉(zhuǎn)寫特征時(shí)如何保證足夠的專家干預(yù)以及設(shè)計(jì)特征時(shí)如何把表面相似現(xiàn)象分開。
[Abstract]:Taking the present reading of 26 Chinese dialects as samples, cluster analysis and principal coordinate analysis were carried out with NTSYS, and the related problems were discussed. The conclusions are as follows: (1) the results of clustering analysis may not be used as a reasonable classification scheme directly; (2) there is no one-to-one relationship between the evolution of the paleo-turbid consonants and the large dialects. (3) the difficulties of constructing the basic feature library of Chinese dialects mainly lie in how to ensure sufficient expert intervention and how to separate the surface similarity phenomenon in the design of features.
【作者單位】： 北京大學(xué)中文系暨中國語言學(xué)研究中心、計(jì)算語言學(xué)教育部重點(diǎn)實(shí)驗(yàn)室;
【基金】：國家社科基金重大項(xiàng)目“基于中國語言及方言的語言接觸類型和演化建模研究”(項(xiàng)目編號(hào):14ZBD102)階段成果
【分類號(hào)】：H17

【相似文獻(xiàn)】

相關(guān)期刊論文 前2條

1 黃行;苗瑤語方言親疏關(guān)系的計(jì)量分析[J];民族語文;1999年03期

2 ;[J];;年期

，

本文編號(hào)：2392072