arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.02073cs.CL
收藏

具有序列到序列模型的语言学启发的形态变化

Linguistically inspired morphological inflection with a sequence to sequence model

CLLE-CNRS(法国国家科学研究中心认知与语言实验室);IRIT-CNRS(法国国家科学研究中心信息研究所);DFKI(德国人工智能研究中心)

Eleni Metheniti, Guenter Neumann, Josef van Genabith

摘要:

屈折变化是人类语言形态学的一个重要组成部分,然而近年来,语言学理论和计算方法的统一却很少。字符串操作的方法被用来推断屈折变化,我们的研究问题是,一个神经网络是否能够学习屈折词素的屈折生产在语言习得的早期阶段,以类似的方式对人类。我们正在使用一个屈折语料库(Metheniti和Neumann,2020)和一个单层seq2seq模型来测试这个假设,其中屈折词缀被学习和预测为一个块,词干被建模为一个字符序列来解释中缀。我们的基于字符语素的模型通过预测词干字符到字符和屈折词缀作为字符块来创建屈折变化。我们进行了三个实验,在给定引理和一组输入和目标特征的情况下创建单词的变形形式,将我们的架构与具有相同超参数,训练和测试集的主流基于字符的模型进行比较。总体而言,对于17种语言,我们注意到在已知词项的屈折变化方面有小幅改进(+0.68%),但我们的模型在预测未知词的屈折变化形式方面的性能稳步提高(+3.7%),在低资源情况下的预测方面也有小幅改进(+1.09%)。

英文摘要:

Inflection is an essential part of every human language's morphology, yet little effort has been made to unify linguistic theory and computational methods in recent years. Methods of string manipulation are used to infer inflectional changes; our research question is whether a neural network would be capable of learning inflectional morphemes for inflection production in a similar way to a human in early stages of language acquisition. We are using an inflectional corpus (Metheniti and Neumann, 2020) and a single layer seq2seq model to test this hypothesis, in which the inflectional affixes are learned and predicted as a block and the word stem is modelled as a character sequence to account for infixation. Our character-morpheme-based model creates inflection by predicting the stem character-to-character and the inflectional affixes as character blocks. We conducted three experiments on creating an inflected form of a word given the lemma and a set of input and target features, comparing our architecture to a mainstream character-based model with the same hyperparameters, training and test sets. Overall for 17 languages, we noticed small improvements on inflecting known lemmas (+0.68%) but steadily better performance of our model in predicting inflected forms of unknown words (+3.7%) and small improvements on predicting in a low-resource scenario (+1.09%)

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。

↑