Complementary Strategies for Low Resourced Morphological Modeling

Erdmann, Alexander; Habash, Nizar

doi:10.18653/v1/w18-5806

Cited by 8 publications

(3 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, it should be noted that, at the time of writing this paper, Modern Greek data sets and text mining approaches are certainly fewer in number when compared against works focusing on other, more widely spoken, popular languages, such as English, German, Spanish, French, Italian, Russian or Chinese. In general, when compared to corresponding tasks in other languages, the obtained accuracy results in Modern Greek web text mining tasks indicate a better overall performance than [50][51][52][53][54][55][56][57], comparable to [58][59][60], or worse than [61][62][63].…”

Section: Approaches On Developing Modern Greek Social Web Text Data Sets and On Modernmentioning

confidence: 98%

The Modern Greek Language on the Social Web: A Survey of Data Sets and Mining Applications

Nikiforos

Voutos

Drougani

et al. 2021

Data

View full text Add to dashboard Cite

Mining social web text has been at the heart of the Natural Language Processing and Data Mining research community in the last 15 years. Though most of the reported work is on widely spoken languages, such as English, the significance of approaches that deal with less commonly spoken languages, such as Greek, is evident for reasons of preserving and documenting minority languages, cultural and ethnic diversity, and identifying intercultural similarities and differences. The present work aims at identifying, documenting and comparing social text data sets, as well as mining techniques and applications on social web text that target Modern Greek, focusing on the arising challenges and the potential for future research in the specific less widely spoken language.

show abstract

Section: Approaches On Developing Modern Greek Social Web Text Data Sets and On Modernmentioning

confidence: 98%

The Modern Greek Language on the Social Web: A Survey of Data Sets and Mining Applications

Nikiforos

Voutos

Drougani

et al. 2021

Data

View full text Add to dashboard Cite

show abstract

“…In future work, modeling semantics can fix such errors, e.g., knowing that rAkb is animate makes plural For future work, we can pre-train on raw corpora to give our model access to such information (Devlin et al, 2019). Indeed Erdmann and Habash (2018) found distributional information to benefit inflectional paradigm clustering in Arabic. Though the benefits should generalize as semantics correlates with inflection class in many languages (Wurzel, 1989;Aronoff, 1992;Harris, 1992;Noyer, 1992;Carstairs-McCarthy, 1994;Corbett and Fraser, 2000;Kastner, 2019).…”

Section: Arabic Error Analysismentioning

confidence: 99%

Frugal Paradigm Completion

Erdmann

Kenter

Becker³

et al. 2020

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

Self Cite

View full text Add to dashboard Cite

Lexica distinguishing all morphologically related forms of each lexeme are crucial to many language technologies, yet building them is expensive. We propose Frugal Paradigm Completion, an approach that predicts all related forms in a morphological paradigm from as few manually provided forms as possible. It induces typological information during training which it uses to determine the best sources at test time. We evaluate our language-agnostic approach on 7 diverse languages. Compared to popular alternative approaches, our Frugal Paradigm Completion approach reduces manual labor by 16-63% and is the most robust to typological variation. * This work was carried out during the first author's internship at Google UK in 2019.

show abstract

“…Erdmann et al (2020) established a baseline for the Paradigm Discovery Problem that clusters the unannotated sentences first by a combination of string similarity and lexical semantics and then uses this clustering as input for a neural transducer. Erdmann and Habash (2018) investigated the benefits of different similarity models as they apply to Arabic dialects. Their findings demonstrated that Word2Vec embeddings significantly underperformed in comparison to the Levenshtein distance baseline.…”

Section: Previous Workmentioning

confidence: 99%

Orthographic vs. Semantic Representations for Unsupervised Morphological Paradigm Clustering

Perkoff¹,

Daniels²,

Palmer³

2021

Proceedings of the 18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

View full text Add to dashboard Cite

This paper presents two different systems for unsupervised clustering of morphological paradigms, in the context of the SIGMOR-PHON 2021 Shared Task 2. The goal of this task is to correctly cluster words in a given language by their inflectional paradigm, without any previous knowledge of the language and without supervision from labeled data of any sort. The words in a single morphological paradigm are different inflectional variants of an underlying lemma, meaning that the words share a common core meaning. They alsousually -show a high degree of orthographical similarity. Following these intuitions, we investigate KMeans clustering using two different types of word representations: one focusing on orthographical similarity and the other focusing on semantic similarity. Additionally, we discuss the merits of randomly initialized centroids versus pre-defined centroids for clustering. Pre-defined centroids are identified based on either a standard longest common substring algorithm or a connected graph method built off of longest common substring. For all development languages, the characterbased embeddings perform similarly to the baseline, and the semantic embeddings perform well below the baseline. Analysis of the systems' errors suggests that clustering based on orthographic representations is suitable for a wide range of morphological mechanisms, particularly as part of a larger system.

show abstract

Complementary Strategies for Low Resourced Morphological Modeling

Cited by 8 publications

References 36 publications

The Modern Greek Language on the Social Web: A Survey of Data Sets and Mining Applications

The Modern Greek Language on the Social Web: A Survey of Data Sets and Mining Applications

Frugal Paradigm Completion

Orthographic vs. Semantic Representations for Unsupervised Morphological Paradigm Clustering

Contact Info

Product

Resources

About