A New Approach to the Supervised Word Sense Disambiguation

Agre, Gennady; Petrov, Daniel; Keskinova, Simona

doi:10.1007/978-3-319-99344-7_1

Cited by 5 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, word embeddings with neural networks are not a cure-all solution for NLP applications (Abraham et al 2018;Agre, van Genabith, and Declerck 2018) for two reasons: the first reason is the necessary text preprocessing, and the second reason is the limitations of word embeddings themselves. Segmenting text into words (i.e., tokenization) is a prerequisite of creating word embeddings (Kudo and Richardson 2018).…”

Section: Nlp Application Typesmentioning

confidence: 99%

Comparison of text preprocessing methods

Chai

2022

Nat. Lang. Eng.

View full text Add to dashboard Cite

Text preprocessing is not only an essential step to prepare the corpus for modeling but also a key area that directly affects the natural language processing (NLP) application results. For instance, precise tokenization increases the accuracy of part-of-speech (POS) tagging, and retaining multiword expressions improves reasoning and machine translation. The text corpus needs to be appropriately preprocessed before it is ready to serve as the input to computer models. The preprocessing requirements depend on both the nature of the corpus and the NLP application itself, that is, what researchers would like to achieve from analyzing the data. Conventional text preprocessing practices generally suffice, but there exist situations where the text preprocessing needs to be customized for better analysis results. Hence, we discuss the pros and cons of several common text preprocessing methods: removing formatting, tokenization, text normalization, handling punctuation, removing stopwords, stemming and lemmatization, n-gramming, and identifying multiword expressions. Then, we provide examples of text datasets which require special preprocessing and how previous researchers handled the challenge. We expect this article to be a starting guideline on how to select and fine-tune text preprocessing methods.

show abstract

Section: Nlp Application Typesmentioning

confidence: 99%

Comparison of text preprocessing methods

Chai

2022

Nat. Lang. Eng.

View full text Add to dashboard Cite

show abstract

“…Two types of embedding compression are supported-the first is implemented by a calculation of the cosine similarity between each embedding vector and the unitary vector (i.e., a vector whose arguments are all equal to 1) of the same dimensionality. The second approach creates the compressed representation of all context words by calculating the cosine similarity between their embedding vectors and the embedding vector of the target word (see [9] for a more detailed description).…”

Section: Representation Of Target and Context Wordsmentioning

confidence: 99%

“…The detailed description of all experiments with the mentioned above data sets are presented in Reference [9]. Table 2 presents only the summary of the best results achieved during the experiments.…”

Section: Comparison With a Knowledge-based Wsd Systemmentioning

confidence: 99%

“…The present paper is devoted to describing a system called WSD Studio, which is intended for the extraction of various features and creating sets of training and test examples for solving the WSD task based on the supervised approach. The main characteristics of the approach applied by the system have been already presented in Reference [9], so in this paper we make more emphasis on the functional architecture of the system, on newly conducted experiments and on deeper discussion of the results. The structure of the paper is as follows: the next section discusses some related work.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Word Sense Disambiguation Studio: A Flexible System for WSD Feature Extraction

2019

Self Cite

View full text Add to dashboard Cite

The paper presents a flexible system for extracting features and creating training and testexamples for solving the all-words sense disambiguation (WSD) task. The system allowsintegrating word and sense embeddings as part of an example description. The system possessestwo unique features distinguishing it from all similar WSD systems—the ability to construct aspecial compressed representation for word embeddings and the ability to construct training andtest sets of examples with different data granularity. The first feature allows generation of data setswith quite small dimensionality, which can be used for training highly accurate classifiers ofdifferent types. The second feature allows generating sets of examples that can be used for trainingclassifiers specialized in disambiguating a concrete word, words belonging to the samepart-of-speech (POS) category or all open class words. Intensive experimentation has shown thatclassifiers trained on examples created by the system outperform the standard baselines formeasuring the behaviour of all-words WSD classifiers.

show abstract

Research on Artificial Neural Networks in Bulgarian Academy of Sciences

Koprinkova–Hristova

2021

Studies in Computational Intelligence

View full text Add to dashboard Cite

A New Approach to the Supervised Word Sense Disambiguation

Cited by 5 publications

References 17 publications

Comparison of text preprocessing methods

Comparison of text preprocessing methods

Word Sense Disambiguation Studio: A Flexible System for WSD Feature Extraction

Research on Artificial Neural Networks in Bulgarian Academy of Sciences

Contact Info

Product

Resources

About