Web-based frequency dictionaries for medium density languages

Kornai, András; Halácsy, Péter; Viktor, Nagy; Oravecz, Csaba; Trón, Viktor; Varga, Dániel

doi:10.3115/1628297.1628298

Cited by 25 publications

(19 citation statements)

References 16 publications

(11 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The corpora are evaluated through the extraction and analysis of wordlists. Other projects that have used web crawls in order to create frequency lists include Kornai et al (2006), who work on Hungarian, and Emerson and O'Neil (2006) on Chinese. A particularly impressive web-derived frequency list is the Google terabyte n-gram collection, made publicly available in 2006 (Brants and Franz 2006).…”

Section: Related Workmentioning

confidence: 99%

The WaCky wide web: a collection of very large linguistically processed web-crawled corpora

Baroni

Bernardini

Ferraresi

et al. 2009

Lang Resources & Evaluation

750

539

View full text Add to dashboard Cite

This article introduces ukWaC, deWaC and itWaC, three very large corpora of English, German, and Italian built by web crawling, and describes the methodology and tools used in their construction. The corpora contain more than a billion words each, and are thus among the largest resources for the respective languages. The paper also provides an evaluation of their suitability for linguistic research, focusing on ukWaC and itWaC. A comparison in terms of lexical coverage with existing resources for the languages of interest produces encouraging results. Qualitative evaluation of ukWaC versus the British National Corpus was also conducted, so as to highlight differences in corpus composition (text types and subject matters). The article concludes with practical information about format and availability of corpora and tools.

show abstract

Section: Related Workmentioning

confidence: 99%

The WaCky wide web: a collection of very large linguistically processed web-crawled corpora

Baroni

Bernardini

Ferraresi

et al. 2009

Lang Resources & Evaluation

750

539

View full text Add to dashboard Cite

show abstract

“…As in Hungarian all the possible constituent orders are grammatical, such a preference could only be attributed to differences in frequency of occurrence. Although no data is available on frequency distributions of different word orders, we made a frequency count of relative pronouns with different case markers available in the Hungarian Webcorpus (Hálacsy et al, 2004; Kornai et al, 2006), showing that the nominative form ‘aki’ is by an order of magnitude more frequent than the accusative form ‘akit’; while frequency differences are smaller for relative pronouns for non-human referents, occurances of nominative forms still dominate (see Table 1). …”

Section: Processing Theories and Their Application To Hungarianmentioning

confidence: 99%

Processing relative clauses by Hungarian typically developing children

Kas

Lukács

2012

Language and Cognitive Processes

View full text Add to dashboard Cite

Hungarian is a language with morphological case marking and relatively free word order. These typological characteristics make it a good ground for testing the crosslinguistic validity of theories on processing sentences with relative clauses. Our study focussed on effects of structural factors and processing capacity. We tested 43 typically developing children in two age groups (ages of 4;11-7;2 and 8;2-11;4) in an act-out task. Differences in comprehension difficulty between different word order patterns and different head function relations were observed independently of each other. The structural properties causing difficulties in comprehension were interruption of main clauses, greater distance between the verb and its arguments, accusative case of relative pronouns, and SO head function relations. Importantly, analyses of associations between working memory and sentence comprehension revealed that structural factors made processing difficult by burdening components of working memory. These results support processing accounts of sentence comprehension in a language typologically different from English.Processing relative clauses is one of the thematic priorities in psycholinguistics. The processing and production of sentences with multiple propositions is a well-suited testing ground for studying the identification and maintenance of thematic roles and their relationships throughout the sentence. There is a good body of results on the topic in English, serving as a basis for different theoretical models on the processing of relative clauses (e. g. Gibson 1998Gibson , 2000MacWhinney, 2000;Lewis et al, 2006;Reali and Christiansen, 2007). The validity of these assumptions for typologically different languages like Hungarian has not been tested yet, though. Experimental research on several languages shows that sentence processing may involve different strategies or different factors in different languages (MacWhinney, 1987;Pléh, 1998;Slobin, 1985). The aim of the present study is to test whether the models based on English are able to explain differences in the processing difficulty of different relative clause structures in Hungarian as well. Examining processing relative clause structures in Hungarian provides an opportunity to test linguistic effects that are not available in typologically different languages like English: 1) being able to vary word order patterns and grammatical functions independently allows us to examine their effects separately, 2) having case-marked relative pronouns makes it possible to test the effect of local cues (i.e. morphological marking) on processing of relative clauses.

show abstract

“…For a full description of the database, see Halácsy et al (2004) and Kornai et al (2006). We included categories that were not associated with each other (either semantically or phonetically) and that were themselves of moderate frequency.…”

Section: Methodsmentioning

confidence: 99%

Interference resolution in retrieval-induced forgetting: Behavioral evidence for a nonmonotonic relationship between interference and forgetting

Keresztes

Racsmány

2012

Mem Cogn

View full text Add to dashboard Cite

Retrieving memories renders related memories less accessible. This phenomenon, termed retrieval-induced forgetting (RIF), is thought to be the result of processes that resolve interference during competitive retrieval. In several studies, researchers have manipulated the level of interference to test different theoretical accounts of RIF (e.g., inhibitory vs. noninhibitory). However, the nature of how interference and RIF are related has not been systematically investigated. Here, we introduce a design that allows for assessing interference during competitive retrieval by measuring the recall RTs associated with target recall. Using such a design, we found that RIF occurred only when interference during competitive retrieval reached moderate levels, but not when it was too low or too high. This finding might indicate that low levels of interference do not trigger interference resolution, whereas interference resolution might fail when the interference reaches extremely high levels.

show abstract

Web-based frequency dictionaries for medium density languages

Cited by 25 publications

References 16 publications

The WaCky wide web: a collection of very large linguistically processed web-crawled corpora

The WaCky wide web: a collection of very large linguistically processed web-crawled corpora

Processing relative clauses by Hungarian typically developing children

Interference resolution in retrieval-induced forgetting: Behavioral evidence for a nonmonotonic relationship between interference and forgetting

Contact Info

Product

Resources

About