语言知识构造对中国日语学习者归纳推论的影响

发布时间：2018-04-22 19:17

本文选题：中国日语学习者 + 语言知识构造　；参考：《清华大学》2015年硕士论文

【摘要】：归纳推论是指根据一部分个别知识推出这些知识共通的一般性规则的思维过程。先行研究通过构建基于大规模语言数据的归纳推论计算模型并进行模拟与实验验证,认为归纳推论的机制可以推测为依存语言呈现出的知识构造。本研究使用基于汉语和日语大规模语言数据统计分析的归纳推论计算模型,将模型的模拟结果运用于以汉语为母语的中国日语学习者的心理实验,探讨在归纳推论中汉语和日语的语言知识构造对外语学习者的影响。本研究设计了汉语和日语2个语言种类的归纳推论调查实验。首先,向汉语和日语各自的计算模型输入相同的8个课题并获取其模拟结果作为调查实验的项目。接着将学习者分组,安排学习者每人都进行2种语言的调查实验。通过比较学习者实验结果与模型的模拟结果,就不同语言的知识构造对中国日语学习者归纳推论的影响可以得出以下结论:首先,学习者在日本的在留时间长度影响其对日语知识构造的习得。在日长期居留的学习者能够更好地掌握日语语言知识构造。实验结果表明,不管实验语言为何,日本在留时间长度在实验结果和计算机模拟结果的相关上,“6个月以上”的学习者的相关高于“6个月以下”的学习者。其次,实验用语的不同对学习者归纳推论的影响并不显著。此次实验调查的参加者多为日语能力水平较高人员,彼此间日语能力差距小,这可能是实验结果差异并不显著的原因之一。今后本研究将完善日语的归纳推论计算模型,通过实施心理学实验验证包含了名词和形容词关系的日语计算模型的可行性。并且,运用改善后的汉语和日语的计算模型,以日语能力不同的学习者为对象进行进一步的调查实验。
[Abstract]:Inductive inference is a process of thinking in which a general rule of common knowledge is derived from some individual knowledge. By constructing a computational model of inductive inference based on large-scale linguistic data and verifying it by simulation and experiment, it is considered that the mechanism of inductive inference can be assumed to be the knowledge structure presented by dependent language. In this study, the model of inductive inference based on the statistical analysis of Chinese and Japanese large-scale language data was used, and the simulation results of the model were applied to the psychological experiments of Chinese native speakers of Japanese. This paper discusses the influence of knowledge structure of Chinese and Japanese on foreign language learners in inductive inference. This study designed an inductive inference investigation of two languages, Chinese and Japanese. First, we input the same 8 subjects to the Chinese and Japanese models and obtain the simulation results as the items of the investigation experiment. Then, the learners were divided into groups and each of them was investigated in two languages. By comparing the experimental results of learners with the simulation results of the model, the following conclusions can be drawn from the influence of the knowledge structure of different languages on the inductive inference of Chinese Japanese learners: first of all, Learners' length of stay in Japan affects their acquisition of Japanese knowledge structure. Learners who stay in Japan for a long time can better master the knowledge structure of Japanese language. The experimental results show that, regardless of the language of the experiment, the correlation between the results of the experiment and the results of computer simulation in Japan is higher than that of the learners of "more than six months" than that of the learners of "less than six months". Secondly, the influence of different experimental terms on learners' inductive inference is not significant. Most of the participants in this experiment are Japanese proficiency, and there is a small gap between them, which may be one of the reasons why the difference is not significant. In the future this study will perfect the Japanese model of inductive inference and verify the feasibility of the Japanese computational model which includes the relationship between nouns and adjectives through psychological experiments. Furthermore, the improved Chinese and Japanese computational models are used to further investigate the learners with different Japanese abilities.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：H36

【相似文献】