Constructing Pseudowords with Constraints on Morphological Features - Application for Polish Pseudonouns and Pseudoverbs

Dołżycka, Joanna; Nikadon, Jan; Formanowicz, Magdalena

doi:10.1007/s10936-022-09884-6

Cited by 3 publications

(2 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dołżycka et al (2022) present a method for generating pseudowords, built upon the foundation of the Wuggy algorithm. Acknowledging, however, the constraints of the algorithm in constructing pseudowords with specific suffixes indicating distinct morphosyntactic categories, they supplement the approach with language-specific rules grounded in real language data.…”

mentioning

confidence: 99%

“…Even though the lower range is technically a negative number, we set it to zero. This is because bigram counts are log-transformed in Num Tool, and the tool does not return negative values.15 An additional optional evaluation step could include native speaker evaluation(Protopapas & Gerakaki 2009, Apostolouda 2018, Soukalopoulou 2021; see alsoDołżycka et al 2022). For instance, speakers can rate the words on a five-point Likert scale (from "definitely not" to "definitely yes") to determine their suitability as Greek words.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Syllable, Bigram, and Morphology-Driven Pseudoword Generation in Greek and a Protocol for Pseudoword Assessment

Apostolouda,

Kosmidis,

Kanetidis

et al. 2023

Preprint

View full text Add to dashboard Cite

Pseudowords are crucial in (psycho)linguistic research, offering a way to study language without meaning interfering. While various methods for pseudoword creation exist, each has its limitations. Traditional approaches create pseudowords by modifying existing words, potentially leading to unintended recognition of the original words. Modern techniques, using high-frequency bigrams or syllable deconstruction, often require linguistic expertise and may not be tailored specifically for Greek, which is our primary focus. Hence, we introduce SyBig (r)- Morph, a novel algorithm for generating Greek pseudowords with unique features that distinguish it from existing pseudoword generation algorithms. SyBig (r)-Morph constructs pseudowords using syllables as the main building block, replicating in this manner the linguistic patterns of Greek words. Furthermore, it is flexible and can be applied to extensive wordlists/databases categorizing syllables by position to ensure phonotactic consistency with user-selected morphosyntactic categories. Another important feature of SyBig (r)-Morph is that it employs dual lexicons to eliminate phonologically invalid combinations. We also outline a specialized protocol for further evaluating the phonological wellformedness of Greek pseudowords, focusing on pseudo-nouns for a production experiment that investigates diverse accentual patterns within and across noun classes. Using an annotated version of the Clean Corpus and the Num Tool, we ensure that these pseudowords adhere to Greek phonotactics and do not closely resemble native words. Selection of essential parameters, including bigram token and type frequency, phonological neighbors, and Levenshtein distance, guarantees accurate phonological well-formedness assessment. This meticulous process results in a curated set of 240 pseudowords for our linguistic experiment.

show abstract

mentioning

confidence: 99%

mentioning

confidence: 99%