Search

Superior stemming algorithms aid significantly in many natural language processing (NLP) applications such as information retrieval. Arabic light-based stemmer is one of the most important stemming algorithms. However, partially due to the highly inflected and complexity of Arabic language morphological structure, most of the existing Arabic light-based stemmer algorithms eliminate a few numbers of suffixes and prefixes or both in the process of recognising the infix patterns to determine roots. The elimination of suffixes and prefixes leads to many inefficient results. Hence, this study aims to develop an improved light-based algorithm of the Arabic stemmer by proposing an appropriate suffixes and prefixes list, which is supported by rules according to word length (without using a morpheme or patterns on a stem). Our improved Dlight Arabic stemmer focuses on determining and removing the infix patterns under many rules on length-words and according to a specific order of the stages of the stemming to extract the double, triple and quadruple roots from long and short Arabic words. To evaluate our proposed light-based Arabic stemmer, we compared our stemmer against existing Arabic stemmers, namely Light10, Condlight and ARLST. The experimental results showed the proposed Develop Arabic Light-Based Stemmer (Dlight) obtained the best performance with 68% of F-measure, while the other three Arabic stemmers yield slightly lower F-measure. Finally, establishing an appropriate list of suffixes and prefixes with word length rules to stem Arabic words can improve the performance of a light-based Arabic stemmer.

This study sought to determine the sequence of L2 morpheme presentation, as well as to determine whether or not the sequence of morpheme presentations correspond with the recognized natural order of morpheme acquisition in English Language Teaching course books utilized with young adult learners at a public sector vocational education institution in Thailand. Qualitative analysis was employed in the scrutinizing of twelve beginner and elementary level ESL and EFL learners course books that have been utilized as the primary teaching material for over a decade by the general education department of the institute. This examination revealed that the morpheme presentation sequence within the selected ELT course books was not analogous with the conclusions in the supporting literature. The findings further indicated that the widely accepted viewpoint of natural order morpheme acquisition was likewise not substantially reflected within the analyzed texts. Albeit, earlier studies have found that an unnatural sequence of morpheme presentation in EFL course books may hamper communicative competence in English, further study is required to establish if this may be a contributing factor for the overall low English proficiency of adult L2 learners in Thailand.

The aim of the present study was to examine the interplay of morphological configuration switching and language switching. The morphological configuration is present in word-formation whenever a word contains more than one free morpheme. The morphological configuration is variable both within and between languages for example in two-digit number names (is the decade named first as in twenty-one or the unit named first as in seventeen) and in compound words (is the modifier or the head named first). In the present experiments, participants had to switch between morphological configurations and between languages (German, English, and Spanish). Language-switch cost was measured as the performance difference between language-switch trials and language-repetition trials. In Experiments 1 and 2, participants performed language-comprehension tasks on two-digit numbers and a language-production task on compound words in Experiment 3. All three experiments revealed an under-additive switch cost pattern in which a larger language-switch cost occurred in morphological configuration-repetition trials than in morphological configuration-switch trials. Thus, the present data indicate integration of the morphological configuration and language into one language-related schema-irrespective of the language task (comprehension vs. production) and the type of stimuli (number words vs. compound nouns). (PsycInfo Database Record (c) 2021 APA, all rights reserved).

The most typical problem in an analysis of natural language is finding synonyms of out-of-vocabulary (OOV) words. When someone tries to understand a sentence containing an OOV word, the person determines the most appropriate meaning of a replacement word using the meanings of co-occurrence words under the same context based on the conceptual system learned. In this study, a word-to-vector and conceptual relationship (Word2VnCR) algorithm is proposed that replaces an OOV word leading to an erroneous morphemic analysis with an appropriate synonym. TheWord2VnCR algorithm is an improvement over the conventional Word2Vec algorithm, which has a problem in suggesting a replacement word by not determining the similarity of the word. After word-embedding learning is conducted using the learning dataset, the replacement word candidates of the OOV word are extracted. The semantic similarities of the extracted replacement word candidates are measured with the surrounding neighboring words of the OOV word, and a replacement word having the highest similarity value is selected as a replacement. To evaluate the performance of the proposed Word2VnCR algorithm, a comparative experiment was conducted using the Word2VnCR and Word2Vec algorithms. As the experimental results indicate, the proposed algorithm shows a higher accuracy than the Word2Vec algorithm.

Sentiment analysis is a classification task where polarity of textual data is identified, i.e. to analyze whether a sentence or document expresses a negative, positive or neutral sentiment. Manipuri is a less privileged, highly agglutinative and tonal language. Despite being a scheduled language of Indian Constitution, it is also a resource constrained language. In this work, we report the sentiment analysis for Manipuri using different types of machine learning based approaches. The dataset used in our work is collected from local daily newspaper. The novelty of this work is that we carry out language specific pre-processing tasks such as transliteration, building negative morpheme based lexicon and filtering of noisy words. Using them as additional linguistic features in our models improves the classification result in terms of precision, recall and F-score. The ensemble voting of best three classifiers based on TF-IDF perform better than BM25 based classifiers and other stand-alone classifiers. Based on this result, we attempt to classify the sentiment of news articles during a certain period of time. Further, we report the finding of deep learning based approaches on the same dataset.

The need to capture intra-word information in natural language processing (NLP) tasks has inspired research in learning various word representations at word, character, or morpheme levels, but little attention has been given to syllables from a syllabic alphabet. Motivated by the success of compositional models in morphological languages, we present a Convolutional-long short term memory (Conv-LSTM) model for constructing Swahili word representation vectors from syllables. The unified architecture addresses the word agglutination and polysemous nature of Swahili by extracting high-level syllable features using a convolutional neural network (CNN) and then composes quality word embeddings with a long short term memory (LSTM). The word embeddings are then validated using a syllable-aware language model ( 31.267 ) and a part-of-speech (POS) tagging task ( 98.78 ), both yielding very competitive results to the state-of-art models in their respective domains. We further validate the language model using Xhosa and Shona, which are syllabic-based languages. The novelty of the study is in its capability to construct quality word embeddings from syllables using a hybrid model that does not use max-over-pool common in CNN and then the exploitation of these embeddings in POS tagging. Therefore, the study plays a crucial role in the processing of agglutinative and syllabic-based languages by contributing quality word embeddings from syllable embeddings, a robust Conv–LSTM model that learns syllables for not only language modeling and POS tagging, but also for other downstream NLP tasks.

A wide variety of mythological elements, cosmological models, rituals, motifs, characters, proverbs, and idioms appear in The Book of Dede Korkut, a unique work and signature of Turkish culture. This study investigates the thematic structure of the last legend of the work, a heroic epic of the Oghuz community. Some structural morphemes, including cosmological structures, thematic contexts, character, and ritual were passed through the filter of analysis. The Gokyay publication of the Dresden copy of The Book of Dede Korkut was taken as the basis for this study, and the final legend (The Story of the Revolt of the Outer Oghuz against the Inner Oghuz and of the Death of Beyrek) in this text is semantically analyzed. The comprehensive analysis of this legend indicates that the phenomenon of opposition developed in the legend is, in fact, not comparable to any legend of the epos in terms of its scope and effect. The series of events that ground the Oghuz tradition or community as reflected in the legend is evaluated through descriptive analysis, and the reason for the hostility between Inner Oghuz and Outer Oghuz is clearly established.

This article deals with the study of Armenian anthroponymic system, particularly, with the morphological structure and etymological semantic of proper names. The aim of this article is to explore derivative models of proprial word formation in order to show the transformation of lexical units and grammatical units into proprial word formation morphemes. The question of interpretation of morphological structure of foreign proper names becomes actual by their rendering in other languages. A common rendering difficulty is caused by almost lack of any explicitly grammatical gender designation in Armenian personal names and surnames. Commen suffixes serve as proprial signals. Key words: proprial nomination, proprial word formation, etymological semantics of onims. DOI: http://dx.doi.org/10.15181/ rh.v0i16.1024

Part-of-speech (POS) tagging is a fundamental task in natural language processing. Korean POS tagging consists of two subtasks: morphological analysis and POS tagging. In recent years, scholars have tended to use the seq2seq model to solve this problem. The full context of a sentence is considered in these seq2seq-based Korean POS tagging methods. However, Korean morphological analysis relies more on local contextual information, and in many cases, there exists one-to-one matching between morpheme surface form and base form. To make better use of these characteristics, we propose a hierarchical seq2seq model. In our model, the low-level Bi-LSTM encodes the syllable sequence, whereas the high-level Bi-LSTM models the context information of the whole sentence, and the decoder generates the morpheme base form syllables as well as the POS tags. To improve the accuracy of the morpheme base form recovery, we introduced the convolution layer and the attention mechanism to our model. The experimental results on the Sejong corpus show that our model outperforms strong baseline systems in both morpheme-level F1-score and eojeol-level accuracy, achieving state-of-the-art performance.

From the point of word formation, the phenomenon of lexical blending is a common productive process, entailing the notion of combination of lexemes in so many languages. In the vast majority of literature on blends, they preserve a linear formation of segments with a shortening of both lexemes. However, in sign languages where morphological categories are mainly encoded by non-concatenative morphology, signed blends can be created by the general mechanism of templatic structures, the combination of lexical bases into a non-linear sequence. Specifically, the main purposes in this study are (i) to provide a comprehensive definition of blending formation in signed modality, (ii) to determine whether there are any structural regularities in the formation of lexical blends in Turkish Sign Language (TA°D), and (iii) to classify TA°D blends according to well-defined criteria. The corpus data to be studied currently include 109 blending formations. Overall, the results demonstrate that TA°D data has familiar properties of blends (named complete blends here) in established spoken languages, as well as modality-specific types of root, simultaneous and initialized blends. We propose a modality-specific categorization, in which blend formation is not limited to linear organization and actual source words.

The clitic morpheme de in Mandarin Chinese has various uses. Typically, it is cliticized to a phrase whether the phrase is nominal or adjectival; it can also occur between two noun phrases when there is no relation of semantic modification. The constructions that involve the latter use of de, known as fake modification constructions, have been theoretically characterized many a time. In the existing characterizations, the morpheme is treated either as a mysteriously inserted lexical item, a modification marker, or a genitive morpheme. The existing accounts suffer from a variety of theoretical and empirical problems. Evidence is presented that in some other constructions and in fake modification constructions, de, while having no lexical semantic content of its own, occupies a position that is otherwise occupied by a two-place predicate. Based on this observation, a partially unitary theoretical account of fake modification constructions is formulated from a parsing perspective in the framework of Dynamic Syntax. In this account, four de-morphemes in fake modification constructions are recognized with different syntactic distributions; however, they all contribute a semantically underspecified predicate that is updated by syntactically constrained or context-based inference.

The article deals with a current problem in the teaching of modern Uzbek literary language in higher education, in particular, philology and language teaching (Uzbek) and linguistics (Uzbek), which has prevailed for a quarter of a century. the existence of different, contradictory, even contradictory interpretations, the negative consequences of which, however, have been criticized for the fact that there have been and are no significant changes in the educational literature. The need to use scientific interpretations of academician A. Khojiev as an acceptable solution to the problem is illustrated by the scientific definition of the morpheme, and it is recommended to teach morphemics in higher linguistics on the basis of this definition, which is an important factor in improving quality and efficiency.

This paper presents a description and analysis of Algonquian verbal classifiers, morphemes that index a salient superordinate feature (such as substance, consistency, or shape) of an argument or no...

Recent research on the acoustic realization of affixes has revealed differences between phonologically homophonous affixes, e.g. the different kinds of final [s] and [z] in English (Plag, Homann & Kunter 2017, Zimmermann 2016a). Such results are unexpected and unaccounted for in widely accepted post-Bloomfieldian item-and-arrangement models (Hockett 1954), which separate lexical and post-lexical phonology, and in models which interpret phonetic effects as consequences of different prosodic structure. This paper demonstrates that the differences in duration of English final S as a function of the morphological function it expresses (non-morphemic, plural, third person singular, genitive, genitive plural, cliticizedhas, and cliticizedis) can be approximated by considering the support for these morphological functions from the words’ sublexical and collocational properties. We estimated this support using naïve discriminative learning and replicated previous results for English vowels (Tucker, Sims & Baayen 2019), indicating that segment duration is lengthened under higher functional certainty but shortened under functional uncertainty. We discuss the implications of these results, obtained with a wide learning network that eschews representations for morphemes and exponents, for models in theoretical morphology as well as for models of lexical processing.

This paper attempts to analyze the Korean sentence classification system. Sentence classification is the task of classifying an input sentence based on predefined categories. However, spelling or space error contained in the input sentence causes problems in morphological analysis and tokenization. This paper proposes a novel approach of Integrated Eojeol (Korean syntactic word separated by space) Embedding to reduce the effect of poorly analyzed morphemes on sentence classification. The paper also proposes two noise insertion methods that further improve classification performance. Our evaluation results indicate that by applying the proposed methods on the existing sentence classifiers, the sentence classification accuracy on erroneous sentences is increased by 8% to 15%.

Morphological analysis of text is one of the most important stages of natural language processing (NLP). Traditional and well-studied problems of morphological analysis include normalization (lemmatization) of a given word form, recognition of its morphological characteristics and their morphological disambiguation. The morphological analysis also involves the problem of morpheme segmentation of words (i.e., segmentation of words into constituent morphs and their classification), which is actual in some NLP applications. In recent years, several machine learning models have been developed, which increase the accuracy of traditional morphological analysis and morpheme segmentation, but performance of such models is insufficient for many applied problems. For morpheme segmentation, high-precision models have been built only for lemmas (normalized word forms). This paper describes two new high-accuracy neural network models that implement morphemic segmentation of Russian word forms with sufficiently high performance. The first model is based on convolutional neural networks and shows the state-of-the-art quality of morphemic segmentation for Russian word forms. The second model, besides morpheme segmentation of a word form, preliminarily refines its morphological characteristics, thereby performing their disambiguation. The performance of this joined morphological model is the best among the considered morpheme segmentation models, with comparable accuracy of segmentation.

Enhanced Conversational Recast (ECR) is an input-based grammatical intervention approach developed from research on statistical learning. Recent research reported evidence demonstrating the efficacy of ECR on the learning of grammatically obligatory morphemes in English-speaking preschool children with developmental language disorder (DLD). This single-case experimental design study, which adopted a within-participant design with single baseline and control item, investigated the efficacy of ECR in promoting the learning of aspect markers in four Cantonese-speaking typically-developing preschool children. Two children demonstrated positive outcomes with the progressive aspect marker ‘ gan2’ given 12 ECR training sessions within a mean dosage of 288. One of these children demonstrated statistically significant gains in the percentage of correct use in the probes. The lack of positive outcomes in the other two children on the earlier developing aspect marker ‘ zo2’ and limitations of the study were discussed. With early evidence established in the typically developing children in this study, future research on Cantonese speaking children with DLD can be considered.

Low-resource languages (LRL) with complex morphology are known to be more difficult to translate in an automatic way. Some LRLs are particularly more difficult to translate than others due to the lack of research interest or collaboration. In this article, we experiment with a specific LRL, Quechua, that is spoken by millions of people in South America yet has not undertaken a neural approach for translation until now. We improve the latest published results with baseline BLEU scores using the state-of-the-art recurrent neural network approaches for translation. Additionally, we experiment with several morphological segmentation techniques and introduce a new one in order to decompose the language’s suffix-based morphemes. We extend our work to other high-resource languages (HRL) like Finnish and Spanish to show that Quechua, for qualitative purposes, can be considered compatible with and translatable into other major European languages with measurements comparable to the state-of-the-art HRLs at this time. We finalize our work by making our best two Quechua–Spanish translation engines available on-line.

Natural language processing is a branch of computer science that combines artificial intelligence with linguistics. It aims to analyze a language element such as writing or speaking with software and convert it into information. Considering that each language has its own grammatical rules and vocabulary diversity, the complexity of the studies in this field is somewhat understandable. For instance, Turkish is a very interesting language in many ways. Examples of this are agglutinative word structure, consonant/vowel harmony, a large number of productive derivational morphemes (practically infinite vocabulary), derivation and syntactic relations, a complex emphasis on vocabulary and phonological rules. In this study, the interesting features of Turkish in terms of natural language processing are mentioned. In addition, summary info about natural language processing techniques, systems and various sources developed for Turkish are given., 8 pages, Natural and Engineering Sciences

Indexed constraints are often used in constraint-based phonological frameworks to account for exceptions to generalizations. A point of contention in the literature on constraint indexation revolves around indexed markedness constraints. While some researchers argue that only faithfulness constraints should be indexed, others argue that markedness constraints should be eligible for indexation as well. This article presents data from Japanese for which a complete synchronic analysis requires indexed markedness constraints but argues that such constraints are only necessary in cases where a phonological repair applies across a morpheme boundary. We then demonstrate that algorithms for learning grammars with indexed constraints can be augmented with a bias towards faithfulness indexation and discuss the advantages of incorporating such a bias, as well as its implications for the debate over the permissibility of indexed markedness constraints.

Dutch features several morphemes with “privative” semantics that occur as left-hand members in compounds (e.g., imitatieleer ‘imitation leather’, kunstgras ‘artificial grass’, nepjuwelen ‘fake jewels’). Some of these “fake” morphemes display great categorical flexibility and innovative adjectival uses. Nep, for instance, is synchronically attested as an inflected adjective (e.g., neppe cupcake ‘fake cupcake’). In this paper, we combine an extensive corpus study of eight Dutch “fake” morphemes with statistical methods in distributional semantics and collexeme analysis in order to compare their semantic and morphological properties and to find out which factors are the driving forces behind their exceptional “extravagant” morphological behavior. Our analyses show that debonding and adjectival reanalysis are triggered by an interplay of two factors, i.e., type frequency and semantic coherence, which allow us to range the eight morphemes on a cline from more schematic to more substantive “fake” constructions.

Word formation, unlike the conjugation of nouns, emerges as a sub-field of morphology that includes the formation of new words and enables word production. In addition to being formed with affixes, new words also occur through combinations of two words coming together. These suffixes (prefix, suffix, covering suffix), which are called word-forming morphemes in morphology, contribute to word production through derivation, which is one of the ways of making words in German. At the same time, formations formed by two independent lexical morphemes, that is, combinations, are another way of making words. These suffixes are added to the first or second independent lexical morphemes in these combinations, and they provide word production and form classes within their own structures. On the other hand, apart from these suffixes, there are prefixoids and suffixoids that are not exactly prefixes or suffixes, but act like them. While the prefix or suffix is dependent on the unit to which it is added, prefixoids and suffixoids are independent lexical morphemes. Although they are independent lexical morphemes, they act as prefixes and suffixes, assuming the same function in words to which they are added. Within the framework determined in the study, only Binary Noun + Noun combinations among the compound words in German were handled. The review is based on the 2011 edition of the Wahrig Deutsches Worterbuch dictionary. For the purpose of the study, foreign words were excluded from the compilation. In the compiled combinations, only the prefixoids were searched, the suffixes were not included. In the study, structural properties of prefixes in Binary Noun + Noun Combinations were examined by using the morphological method.

Using Deep Neural Networks (DNN) as an advanced Artificial Neural Networks (ANN) has become widespread with the development of computer technology. Although DNN has been applied for solving Automatic Speech Recognition (ASR) problem in some languages, DNN-based Turkish Speech Recognition has not been studied extensively. Turkish language is an agglutinative and a phoneme-based language. In this study, a Deep Belief Network (DBN) based Turkish phoneme and speech recognizer is developed. The proposed system recognizes words in the system vocabulary and phoneme components of out of vocabulary (OOV) words. Sub-word (morpheme) based language modelling is implemented into the system. Each phoneme of Turkish language is also modelled as a sub-word in the model. Sub-word (morpheme) based language model is widely used for agglutinative languages to prevent excessive vocabulary size. The performance of the suggested DBN based ASR system is compared with the conventional recognition method, GMM (Gaussian Mixture Method) based Hidden Markov Model (HMM). Regarding to performance metrics, the recognition rate of Turkish language is improved in compare with previous studies.

This study was a morphology descriptive analysis of Manggarai language (BM) which was focused on three areas those are language typology based on the level of morphological use namely analytic, synthetic, and polysynthetic; morpheme identification; and morphological process. Therefore the aim of this study was to describe them. Data used in this study were library data which were collected using simak method and introspection method. The data were analyzed using agih method. Besides, there are three main results of the analysis data in this study. The first showed that BM is the analytic language because morphologically words in BM are monomorphemic. The second finding showed that bound morphemes in BM not related to affixation, but they are related to the combination of synonyms and clitictation. The last finding found that there is no affixation in the morphological process, but it is only compound words and reduplication.

A partir da análise para formas nominais truncadas e para blends em português brasileiro sugeridas em Scher (2011, 2013, 2016, 2018) e Marangoni Junior (2020), respectivamente, que se valem da presença de um morfema avaliativo [EVAL] na estrutura sintática, discute-se a relevância desse morfema como um traço pertinente ao repositório de traços morfossintáticos universais, considerando-se, principalmente, as observações sobre o estatuto de morfemas avaliativos em duas propostas distintas, nomeadamente, Villalva (2000) e Prieto (2005). As análises descritas no texto, baseadas em uma abordagem sintática dos processos de formação de palavras, evidenciam a configuração do morfema avaliativo em posição de adjunção. Além disso, a presença desse morfema codifica instruções específicas para as interfaces de PF e LF.

The features of the formation of an adherent negative assessment of neologism on the material of the headlines of electronic media are considered. The results of structural-semantic, contextual and pragmatic analysis of media neologisms expressing an adherent negative assessment are presented. The analysis of the main methods of forming an adherent negative assessment of neologisms under the influence of narrow (title context) and wide (context of an integral media product) verbal contexts, as well as under the influence of extra-linguistic conditions of use (communicative-pragmatic settings of the author, features of the subject of evaluation, specificity of the value system). It is concluded that the adherent negative evaluation of the neologisms is formed in the context due to the updating and intensification of the negative evaluation components of the semantics of the producing word, at the same time, due to the conditions of use, it is possible to de-intensify the estimated semantics of the structural components of the neologism, up to the neutralization of the evaluation of the neologism, as well as reversal (change of sign) of the assessment from positive to negative. Cases of the formation of adherent negative evaluation semantics in neologisms created on the basis of evaluation-neutral components are also analyzed. The practical significance of the study lies in the possibility of applying its results in journalistic practice, as well as in the practice of teaching university courses in morphemic, lexicology and special courses on the study of the language of the media, pragmalinguistics.

This study examined how lexical properties, such as word frequency, word length, and morphological features, affect the word recognition of Korean Hangul among adult readers. Ninety-four native Korean students performed a lexical decision task on disyllabic and trisyllabic words and nonwords. Results of cross-classified and hierarchical linear modeling showed a significant frequency effect but null effects of word length and their interactions on word recognition. The null syllable effect might be contributable to the block structure of syllables, which does not require syllabic decomposition in recognition. Given that Chinese-derived Sino-Korean words account for about 70% of the Korean lexicon, morphological features embedded in the word were examined in order to further understand the morphological undercurrent of multisyllabic word recognition. Word type (bound-morphemic words vs. compound words), word origin (native or semi-native words vs. Chinese-derived words), and morphemic transparency (opaque morpheme vs. transparent morpheme) were significant predictors of the speed of visual word recognition. An analysis of frequency split showed that compound words and morpheme transparency facilitated word recognition for high frequency words, while the number of morphemes within the word, Korean native and semi-native words, and morphemic transparency facilitated word recognition for low frequency words. These results suggest that high and low frequency words function differently according to morphological information available within the multisyllabic word in visual word recognition in Hangul.

Common tests of morphological awareness measure both morphology and syntax by requiring participants to fit words and pseudowords into sentences by adding or removing affixes. We report the results of a study testing a new word level task. College students viewed transparent words (without phonological or orthographic shifts) and used a keyboard to indicate whether the items contained 1, 2, 3, or 4 morphemes. Morpheme counting accuracy was strongly and significantly correlated with sentence level tests of morphological awareness, also grouping with them in a factor analysis, suggesting that the tasks measure a similar construct. Morpheme counting accuracy was also strongly and significantly correlated with the word identification and passage comprehension measures from the WJ-IV. Crossed random-effects modeling showed that all tasks were sensitive to word frequency and vocabulary. However, different MA tasks varied in their sensitivity to the sublexical properties of words. Responses in the sentence level tasks were sensitive to word and bigram frequency while responses in the word level task was sensitive to base frequency and the number of morphemes. Our findings suggest that conscious knowledge of root words and affixes can be directly measured at the word level without a syntactic component to the task and that responses do capture variation in the ability to decompose complex words into their component morphemes.

Automatic question generation facilitates the smart assessment for the evaluator to assess the student skills. Several methods were proposed to generate distractors for non-factoid cloze question using different similarity measures. This study presents a method for automatic generation of affix based distractor for Tamil fill-in-the-blank questions which are mainly used for learning Tamil grammar morphological details and vocabulary. In this study, affix based distractor generation is proposed as two step pipelined process: 1) Distractor candidate collection: This generation mainly relies on certain regularities manifest in high dimensional spaces which implicitly hybrids the orthographic and semantic features. 2) Distractor filtering: Filtering is trained as Learning-to-Rank models to persist the reliability in distractor generation. Feature based Listwise approaches (ListNet and ListMLE) were used which uses caserole relationship, subject-verb agreement, POS tag in addition to similarity measures. Experiments were done with annotated dataset (TamilMCQs) taken from 5th to 12th grade Tamil text books. Experimental results show that hybridization of spelling and semantic features highly improves the plausibility in distractor generation and then, ListMLE method improves the reliability in distractor generation compared to ListNet method. As a overall, our proposed pipelined process increases plausibility and reliability in distractor generation.

Previous studies have shown that English speakers use a range of factors including locality, information structure, and semantic parallelism to interpret clausal ellipsis structures. Yet, the relative importance of each factor is currently underexplored. As cues to information structure and semantic parallelism are often implicit in English, we turned to Persian which marks information structure overtly via word order scrambling and uses the -rā morpheme to indicate definiteness/specificity on direct objects. To determine what strategies Persian speakers use to disambiguate clausal ellipsis, we conducted a naturalness rating study and sentence completion task on polarity stripping structures. Our results show that information structure and parallelism strongly influence correlate resolution in both tasks, but that a weaker preference for a local correlate emerges in scrambling in the sentence completion task. As these results diverge from those obtained in English studies, we speculate that the morphosyntactic properties of Persian constrain the strategies the processer uses in selecting a contrastive correlate and resolving ambiguity in stripping ellipsis.

The article describes a multilingual linguistic database for Turkic languages, which is an integral model that includes taxonomic and situational ontologies. The basis of necessity for the development is the fact that the proposed linguistic database is a resource base for a whole set of linguistic processors. This is especially important for the Turkic languages since most of them belong to the type of low-resource languages, precisely because of the lack of linguistic bases, especially of the semantic-syntactic level. At the same time, the proposed model of the linguistic database developed with due regard to the structural and functional features of the Turkic languages, which will increase the efficiency of the linguistic software.

While the theoretical models of morphological processing in Roman alphabets indicate prelexical activation, a model established in Korean suggests postlexical activation. To extend the model of Korean morphological processing, this study examined within-scriptal (Hangul-Hangul prime-target pairs) and cross-scriptal (Hanja-Hangul prime-target pairs) priming effects on the recognition of Sino-Korean compound words in Hangul as a function of adult readers’ Hanja proficiency using priming lexical decision tasks. Experiment 1 (n = 54) examined the constituent morphemic effects of Hangul and Hanja primes, while Experiment 2 (n = 67) investigated morphemic decomposition in Hangul and Hanja primes. Participants with skilled Hanja proficiency showed robust constituent morphemic effects with both Hangul and Hanja primes in isolation, while less-skilled participants did not show the effects with Hangul and Hanja primes (Experiment 1). The skilled group showed efficient morphological decomposition in both within- and cross-script conditions. However, the less-skilled group did not show morphological effects in the cross-script condition but the same effect in the within-script condition (Experiment 2). The skilled group showed ortho-phonological inhibitory effects on Hangul recognition resulting from competitions among activated neighbors, but the less-skilled counterpart did not show the effect. Based on the findings of this study, two differing pathway models of morphological processing in Hangul are proposed for readers with different Hanja proficiency. Morphological processing in Sino-Korean compound words seems to be prelexical for skilled readers, whereas it is postlexical for less-skilled readers.

This paper aims examines the role of hierarchical inference in sound change. Through hierarchical inference, a language learner can distribute credit for a pronunciation between the intended phone and the larger units in which it is embedded, such as triphones, morphemes, words and larger syntactic constructions and collocations. In this way, hierarchical inference resolves the longstanding debate about the unit of sound change: it is not necessary for change to affect only sounds, or only words. Instead, both can be assigned their proper amount of credit for a particular pronunciation of a phone. Hierarchical inference is shown to generate novel predictions for the emergence of stable variation. Under standard assumptions about linguistic generalization, it also generates a counterintuitive prediction of a U-shaped frequency effect in an advanced articulatorily-motivated sound change. Once the change has progressed far enough for the phone to become associated with the reduced pronunciation, novel words will be more reduced than existing words that, for any reason, have become associated with the unreduced variant. Avoiding this prediction requires learners to not consider novel words to be representative of the experienced lexicon. Instead, learners should generalize to novel words from other words that are likely to exhibit similar behavior: rare words, and the words that occur in similar contexts. Directions for future work are outlined.

Two major problems of Korean part-of-speech (POS) tagging are that the word-spacing unit is not mapped one-to-one to a POS tag and that morphemes should be recovered during POS tagging. Therefore, this article proposes a novel two-step Korean POS tagger that solves the problems. This tagger first generates a sequence of lemmatized and recovered morphemes that can be mapped one-to-one to a POS tag using an encoder-decoder architecture derived from a POS-tagged corpus. Then, the POS tag of each morpheme in the generated sequence is finally determined by a standard sequence labeling method. Since the knowledge for segmenting and recovering morphemes is extracted automatically from a POS-tagged corpus by an encoder-decoder architecture, the POS tagger is constructed without a dictionary nor handcrafted linguistic rules. The experimental results on a standard dataset show that the proposed method outperforms existing POS taggers with its state-of-the-art performance.

This study examined the use of non-standard English features in 16-year-old students’ writing in the UK. Adopting a diachronic approach to the investigation of students’ written production, the stu...

The relevance of the article is in introduction of the results of experimental linguistic studies concerning professional land manager’s basic vocabulary morphemes nesting on practice. The purpose of the article is to describe the experiment of involving the statistic word-building nests analysis into communicative practice while studying professionally oriented English to improve soft foreign language communicative skills of future land managers. The conclusion of the study is that land manager’s special vocabulary is characterized by some morphemes nesting to be selected by students themselves for the shortening of the term list for further dialogues within the positive linguistic studies experiment. The basic morphemes (without morpheme land-) include such representatives as adjust-, aer-, altitude-, area-, boni-, cadastr-, cartograph-, cost-, determination-, direction-, earth/Earth-, field-, fix-, geo-, grid-, lease-, level-, locat-, longitude-, map-, measure-, mone-, position-, photo-, relief-, soil-, survey-, terrain-, topograph-, work. The most spread types of land manager’s vocabulary concerning basic morphemes are attribute-noun two components word combinations, noun-noun two components word combination and attribute-noun three-four component word combinations. Besides, the most spread basic morphemes within land manager’s vocabulary are map-, geo-, soil-, measure- and direction-. The prospects of the research is the additional detailed characteristics of the morpheme land- as the most important for land manager’s professional vocabulary morpheme for communicative purposes.

Previous research suggests that while free morpheme identification during visual word recognition is position-independent, suffixes are activated only when they occur after the stem. Surprisingly, prefix position coding has not yet been assessed. This point is important given that some experimental studies demonstrated clear processing differences between prefixes and suffixes. In this study we examined whether Spanish suffixes and prefixes are recognized independently of their position by adapting the Crepaldi, Rastle, and Davis’s (2010) experimental paradigm. We observed that morphologically structured nonwords in which the affix occurs in its typical position (e.g., curiosura, disgrave) are rejected more slowly and less accurately than their matched orthographic controls (e.g., curiosula, dusgrave). Crucially, such morpheme interference effect is completely absent when the morphemes are inverted (i.e., uracurios and gravedis are rejected as easily as ulacurios and gravedus). Our data provide strong support to the hypothesis that all affix processing is sensitive to position.

Chinese coordinative compound words are common and unique in inter-character semantic and orthographic relationships. This study explored the inter-character orthographic similarity effects on the recognition of transparent two-morpheme coordinative compound words. Seventy-two native Chinese readers participated in a lexical decision task. The findings demonstrated robust inhibitory inter-character orthographic similarity effects, intra-word character reversal effects, and inter-character semantic similarity effects. These results were compared to those of previous studies on coordinative compound word recognition and on the orthographic similarity phenomenon at both character and word levels. The findings were explained with the multi-level representational model of morphological processing of Chinese compound words (Zhou and Marslen-Wilson in Psychologia 43:47-66, 2000). The model was further extended by adding the activation of morpho-orthographic relationships and the mapping of morphemic orthographic information onto the semantic information of both morphemes and whole words.

Do the cognitive benefits of interleaving—the method of alternating between two or more skills or concepts during training—extend to foreign language learning? In four experiments, we investigated the efficacy of interleaved versus conventional blocked practice for teaching adult learners to conjugate Spanish verbs in the preterite and imperfect past tenses. In the first two experiments, training occurred during a single session and interleaving between tenses began during the presentation of introductory content (Experiment 1) or during randomly ordered verb conjugation practice trials at the end of the training session (Experiment 2). This yielded, respectively, numerically higher performance in the blocked group and equivalent performance in the interleaved and blocked groups on a 2-day delayed test. In Experiments 3 and 4, the amount of training was increased across 2 weekly sessions in which the blocked group trained on 1 tense per session and the interleaved group trained on both tenses per session, with random interleaving occurring during verb conjugation practice trials. Interleaving yielded substantially better performance on a 1-week delayed test. Thus, although interleaving did not confer an advantage over blocking under 2 different single-session training schedules, it improved learning when used to practice conjugating verbs across multiple training sessions. These results constitute the first demonstration of an interleaving effect for foreign language learning. (PsycINFO Database Record (c) 2019 APA, all rights reserved)

This paper presents an automatic method for extracting, processing, and analysis of customer opinions on Arabic social media. We present a four-step approach for mining of Arabic tweets. First, Natural Language Processing (NLP) with different types of analyses had performed. Second, we present an automatic and expandable lexicon for Arabic adjectives. The initial lexicon is built using 1350 adjectives as seeds from processing of different datasets in Arabic language. The lexicon is automatically expanded by collecting synonyms and morphemes of each word through Arabic resources and google translate. Third, emotional analysis was considered by two different methods; Machine Learning (ML) and rulebased method. Finally, Feature Selection (FS) is also considered to enhance the mining results. The experimental results reveal that the proposed method outperforms counterpart ones with an improvement margin of up to 4% using F-Measure.

This paper presents an efficient method to build a corpus to train natural language understanding (NLU) modules. Conventional corpus creation methods involve a common cycle: a subject is given a specific situation where the subject operates a device by voice, and then the subject speaks one utterance to execute the task. In these methods, many subjects are required in order to build a large-scale corpus, which causes a problem of increasing lead time and financial cost. To solve this problem, we propose to incorporate a “probing question” into the cycle. Specifically, after a subject speaks one utterance, the subject is asked to think of alternative utterances to execute the same task. In this way, we obtain many utterances from a small number of subjects. An evaluation of the proposed method applied to interview-based corpus creation shows that the proposed method reduces the number of subjects by 41% while maintaining morphological diversity in a corpus and morphological coverage for user utterances spoken to commercial devices. It also shows that the proposed method reduces the total time for interviewing subjects by 36% compared with the conventional method. We conclude that the proposed method can be used to build a useful corpus while reducing lead time and financial cost.

This corpus-based study provides a baseline of complex word usage patterns in the spontaneous speech of English preschool children to ascertain the characteristics of their derivative vocabulary before literacy development affects language skills. Frequencies of suffixed derivatives produced by ( N = 243) children aged 2–5 and their caregivers were extracted for 58 suffix variants, yielding 558 types from the former and 1,364 from the latter. Between the youngest and oldest groups, 11 suffix categories increased significantly in type frequency, compared with 22 in the caregiver dataset. All derivative types were classified for transparency of meaning and simplicity of form on a five-point analysability scale. Around 59% of both the child and caregiver derivative vocabulary sets were classified as transparent regardless of age, suggesting that the potential analysability of the preschool child’s input remains surprisingly invariant over time. The study provides baseline data for future studies on the development of morphological awareness in English-speaking schoolchildren.

Since the 1970s, findings from Morpheme Order Studies (MOS) have suggested that the emergence of morphemes follows a predictable order in L2 English. In this paper we show how the tools and practices in Learner Corpus Research (LCR) offer a richer descriptive basis, which is achieved with Interlanguage Annotation (ILA), a manual, fined-grained, purpose-oriented type of annotation. Additionally, we use a standardised placement test, since proficiency level has been overlooked in most previous MOS. Both of these practices provide a more detailed description of morpheme accuracy order across different levels. We analyse four proficiency levels (A1-B2) in a subcorpus of L1 Spanish-L2 English secondary-school learners from the CORpus of English as a Foreign Language (COREFL). Our results partially confirm findings from previous MOS, but also reveal key findings that had gone previously unnoticed regarding the role of proficiency level and the subtype of errors, which are relevant factors for SLA research.