基于分布格模型的英语词类习得研究
发布时间：2018-11-27 17:01
【摘要】：本文采用新型的词类学习模型—分布格学习模型,考察分布特征对词类习得的影响。分布格是以词的分布特征相似性为基础的一种分析模型。本文选取英语儿童语料库Brent中100个高频词和500个高频语境作为学习对象,通过该模型计算100个高频词的分布特征相似性并进行分类,生成一系列词类。研究发现,分布格学习模型可以准确地生成名词、动词、形容词、代词、副词等一系列常见的词类,总体准确率为90%,其中动词的准确率高达94%,进一步验证分布特征在词类形成中发挥着重要作用。
[Abstract]:In this paper, a new type of part of speech learning model, the distributed lattice learning model, is used to investigate the effect of distribution characteristics on the acquisition of parts of speech. Distribution case is an analytical model based on the similarity of word distribution characteristics. This paper selects 100 high-frequency words and 500 high-frequency context in the English children's corpus Brent as the learning objects, calculates the similarity of the distribution characteristics of the 100 high-frequency words and classifies them through the model, and generates a series of parts of speech. It is found that the distributed case learning model can accurately generate a series of common parts of speech, such as nouns, verbs, adjectives, pronouns, adverbs, etc. Further verify that the distribution features play an important role in the formation of parts of speech.
【作者单位】： 北京航空航天大学;伦敦国王学院;
【分类号】：H319
本文编号：2361492
[Abstract]:In this paper, a new type of part of speech learning model, the distributed lattice learning model, is used to investigate the effect of distribution characteristics on the acquisition of parts of speech. Distribution case is an analytical model based on the similarity of word distribution characteristics. This paper selects 100 high-frequency words and 500 high-frequency context in the English children's corpus Brent as the learning objects, calculates the similarity of the distribution characteristics of the 100 high-frequency words and classifies them through the model, and generates a series of parts of speech. It is found that the distributed case learning model can accurately generate a series of common parts of speech, such as nouns, verbs, adjectives, pronouns, adverbs, etc. Further verify that the distribution features play an important role in the formation of parts of speech.
【作者单位】： 北京航空航天大学;伦敦国王学院;
【分类号】：H319
【相似文献】
相关期刊论文 前2条
1 章莹;关于萨图克·布格拉汗麻扎的研究[J];西域研究;1996年01期
2 ;[J];;年期
相关会议论文 前1条
1 朱自强;黄国祥;李新剑;;用分形分析确定最佳重力布格密度的方法探讨[A];1993年中国地球物理学会第九届学术年会论文集[C];1993年
相关重要报纸文章 前1条
1 吴刚 罗大森;鲁布格派出所调解群体纠纷保稳定[N];黔西南日报;2009年
相关博士学位论文 前1条
1 李恩;布格呋喃与CYP450/P-糖蛋白的相互作用及生物学效应[D];中国协和医科大学;2010年
，本文编号：2361492
本文链接：https://www.wllwen.com/waiyulunwen/yingyulunwen/2361492.html
