RECORD
Expanding Abbreviations in a Strongly Inflected Language: Are Morphosyntactic Tags Sufficient?
Computation and Language · Submitted 20 Aug 2017
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Evidence is presented that the correct inflected form of an expanded abbreviation can in many cases be deduced solely from the morphosyntactic tags of the context.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
In this paper, the problem of recovery of morphological information lost in abbreviated forms is addressed with a focus on highly inflected languages. Evidence is presented that the correct inflected form of an expanded abbreviation can in many cases be deduced solely from the morphosyntactic tags of the context. The prediction model is a deep bidirectional LSTM network with tag embedding. The training and evaluation data are gathered by finding the words which could have been abbreviated and using their corresponding morphosyntactic tags as the labels, while the tags of the context words are used as the input features for classification. The network is trained on over 10 million words from the Polish Sejm Corpus and achieves 74.2% prediction accuracy on a smaller, but more general National Corpus of Polish. The analysis of errors suggests that performance in this task may improve if some prior knowledge about the abbreviated word is incorporated into the model.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | In this paper, the problem of recovery of morphological information lost in abbreviated forms is addressed with a focus on highly inflected languages. | background 0.83 | 0.61 | 0.04 | asserted 0.60 |
| s1 | Evidence is presented that the correct inflected form of an expanded abbreviation can in many cases be deduced solely from the morphosyntactic tags of the context. | result 0.95 | 0.88 | 0.84 | asserted 0.86 |
| s2 | The prediction model is a deep bidirectional LSTM network with tag embedding. | method 1.00 | 0.84 | 0.04 | asserted 1.00 |
| s3 | The training and evaluation data are gathered by finding the words which could have been abbreviated and using their corresponding morphosyntactic tags as the labels, while the tags of the context words are used as the input features for classification. | method 1.00 | 0.93 | 0.03 | asserted 0.99 |
| s4 | The network is trained on over 10 million words from the Polish Sejm Corpus and achieves 74.2% prediction accuracy on a smaller, but more general National Corpus of Polish. | result 0.98 | 0.96 | 0.09 | asserted 1.00 |
| s5 | The analysis of errors suggests that performance in this task may improve if some prior knowledge about the abbreviated word is incorporated into the model. | future work 0.58 | 0.85 | 0.13 | hedged possibility 1.00 |
What was examined, and what was not
- arXiv listing abstract v2
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
In this paper, the problem of recovery of morphological information lost in abbreviated forms is addressed with a focus on highly inflected languages. Evidence is presented that the correct inflected form of an expanded abbreviation can in many cases be deduced solely from the morphosyntactic tags of the context. The prediction model is a deep bidirectional LSTM network with tag embedding. The training and evaluation data are gathered by finding the words which could have been abbreviated and using their corresponding morphosyntactic tags as the labels, while the tags of the context words are used as the input features for classification. The network is trained on over 10 million words from the Polish Sejm Corpus and achieves 74.2% prediction accuracy on a smaller, but more general National Corpus of Polish. The analysis of errors suggests that performance in this task may improve if some prior knowledge about the abbreviated word is incorporated into the model.
Human review
How the literature uses it
What this paper does with the arXiv papers it cites
What it cites · 1 sentence without a confident label
Uses the method or tool of 1 paper
-
The training was performed using the error backpropagation algorithm [ref] with cross-entropy criterion and Adam optimizer [ref] with the same settings of learning rate and momentum parameters as suggested in the paper.
Adam: A Method for Stochastic Optimization · Experimental setup · 0.99
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.