Seq2seq for Morphological Reinflection: When Deep Learning Fails

Senuma, Hajime; Aizawa, Akiko

doi:10.18653/v1/k17-2011

Cited by 5 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also experimented with a monolingual MED model, but accuracy was close to zero for all languages. This was an expected outcome, given earlier results (Senuma and Aizawa 2017). We do, thus, not discuss this baseline in the remaining parts of the paper.…”

Section: Baselinesmentioning

confidence: 72%

Learning to Learn Morphological Inflection for Resource-Poor Languages

Kann

Bowman

Cho

2020

AAAI

View full text Add to dashboard Cite

We propose to cast the task of morphological inflection—mapping a lemma to an indicated inflected form—for resource-poor languages as a meta-learning problem. Treating each language as a separate task, we use data from high-resource source languages to learn a set of model parameters that can serve as a strong initialization point for fine-tuning on a resource-poor target language. Experiments with two model architectures on 29 target languages from 3 families show that our suggested approach outperforms all baselines. In particular, it obtains a 31.7% higher absolute accuracy than a previously proposed cross-lingual transfer model and outperforms the previous state of the art by 1.7% absolute accuracy on average over languages.

show abstract

Section: Baselinesmentioning

confidence: 72%

Learning to Learn Morphological Inflection for Resource-Poor Languages

Kann

Bowman

Cho

2020

AAAI

View full text Add to dashboard Cite

show abstract

“…Another GRU network is deployed as a decoder to generate the inflection. The UTNII 2017 model, published in [15] is based on the seq2seq model, and with its configuration, it was the second best of 2017 in the high-resource scenarios. The Hamburg 2018 model published by Schroder [16] introduces the concept of patches that act as string transducer actions.…”

Section: Related Surveymentioning

confidence: 99%

B-Morpher: Automated Learning of Morphological Language Characteristics for Inflection and Morphological Analysis

Kovács

Szabó

2022

Cybernetics and Information Technologies

View full text Add to dashboard Cite

The automated induction of inflection rules is an important research area for computational linguistics. In this paper, we present a novel morphological rule induction model called B-Morpher that can be used for both inflection analysis and morphological analysis. The core element of the engine is a modified Bayes classifier in which class categories correspond to general string transformation rules. Beside the core classification module, the engine contains a neural network module and verification unit to improve classification accuracy. For the evaluation, beside the large Hungarian dataset the tests include smaller non-Hungarian datasets from the SIGMORPHON shared task pools. Our evaluation shows that the efficiency of B-Morpher is comparable with the best results, and it outperforms the state-of-theart base models for some languages. The proposed system can be characterized by not only high accuracy, but also short training time and small knowledge base size.

show abstract

Automated Learning of Hungarian Morphology for Inflection Generation and Morphological Analysis

Szabó

Kovács

2020

IJEEI

View full text Add to dashboard Cite

Seq2seq for Morphological Reinflection: When Deep Learning Fails

Cited by 5 publications

References 14 publications

Learning to Learn Morphological Inflection for Resource-Poor Languages

Learning to Learn Morphological Inflection for Resource-Poor Languages

B-Morpher: Automated Learning of Morphological Language Characteristics for Inflection and Morphological Analysis

Automated Learning of Hungarian Morphology for Inflection Generation and Morphological Analysis

Contact Info

Product

Resources

About