Verb Sequences Research Articles

Chinese question answering pattern learning based on self-training mechanism and Web

In the past,the learning for QA pattern relies on the labeled data,and the definition of pattern and the scoring method for the candidate answers are over simplified.The verb and noun sequence was extracted as the skeleton pattern to expand definition of QA pattern.In the learning process,a learning mechanism was established based on self-training.At first,the initial study was completed on a labeled QA pair,then the system would automatically select the reliable data for self training through searching in the Web while the system was running.The scoring method of the candidate answers was also improved by applying several heuristic rules.The experimental results show that the performance of Chinese QA system based on our method is improved significantly.

Additional facts about noun incorporation (in Inuktitut)

This article examines a range of data concerning noun incorporation in Inuktitut. A number of facts are known in the linguistic literature of Inuit languages but are often not addressed in noun incorporation analyses. Possessor stranding is not permitted in noun incorporation, with the exception of a subclass of incorporating verbs which describe location and direction. Names are readily incorporated, a fact which poses a problem for analyses claiming that incorporated nominals are predicates. New empirical evidence about noun incorporation in Inuktitut is also introduced. This includes the fact that noun incorporation verbs can sometimes be optionally transitive, as well as the fact that passive morphemes may be added to the incorporated noun plus verb sequences. Throughout the paper it is shown that all the facts are accounted for by a Root Movement analysis. The nominal is incorporated not because it is a nominal but because it is a root and there is no verbal root element in this construction. Another concern discussed is the translation of elements within complex words, using English glosses. The paper concludes that more empirical research is required on languages which display noun incorporation.

When I started to using BLUR

This article introduces BLUR, an electronic corpus of Blues lyrics from the early twentieth century, as a valuable resource for the study of syntactic phenomena in Earlier African American English and investigates the properties and origins of a peculiar construction found in it. The design of the BLUR corpus is presented, and methodological consequences resulting from the nature of the texts are discussed. Subsequently, a few noteworthy syntactic structures documented in it are briefly illustrated. The most challenging of these, a sequence of inchoative verbs (notably begin, start, or commence), the particle to, and a verbal –ing form, as in begin to falling, is then analyzed in some detail, considering syntactic constraints and historical documentation in potentially related varieties, represented by electronic corpora and other sources from Late Middle English to present-day dialects. Based on these findings, a hypothesis on the likely origins and diffusion of this construction is developed.

Variable deletion of French ne: a cross-stylistic perspective

This paper considers the socio-stylistic distribution of the French variable morpho-syntactic particle ne. The interspeaker axes of variation in ne are summarised first, prior to a consideration of some intraspeaker data deriving from a corpus of spoken French. As well as presenting group scores, we examine intraspeaker variation in ne by focusing on use of the variable by a single speaker in both speech styles recorded. One principal finding is that the formal speech situation was such as to work against the strong constraint that induces French speakers to produce clitic+verb sequence as perhaps the default form; a constraint that strongly favours ne deletion. In informal style, several examples of retention of ne seem to show that the tone of a stretch of discourse can vary independently of the informality of the speech style. Several of the stylistic effects produced by the informants in conversation style through their use of ne show that the micro-style variation which is present in both of the broad styles is reflected in the use of ne through a reduction in degrees of style shift, since the ‘formal’ episodes in informal style call for the use of ne frequently relative to formal style — albeit within the context of very little use of ne overall.

The Effects of Case Marking Information on Korean Sentence Processing

Two experiments were conducted to determine whether a parser's on-line processes are affected by manipulation of case markings in Korean sentence processing, and to evaluate the usefulness of a Ranked Flagged Serial Parser (RFSP) model in predicting Korean ambiguity resolution processes. By using a self-paced word-by-word reading task, in Experiment 1, relative and coordinate sentences of the form NP-ka (nominative), NP-lul (accusative), and Verb were compared to those sentences of the form NP-nun (topic marker), NP-lul, and Verb. In Experiment 2, relative sentences that have NP-ka, NP-ka, and Verb sequence were compared to those sentences of NP-nun, NP-ka, and Verb sequence. In addition, reversed relative clause sentences were compared after changing case markings of the last NP. The topic marker -nun changed overall patterns of word-by-word reading times of the sentences in both experiments. The results clearly showed that local grammatical cues, such as case markings, are used in generating syntactic hypotheses. We also found that the RFSP model successfully predicted local processing loads and reparsing difficulties in Korean sentence comprehension. Suggestions for further studies were made.

Toward a typology of directional motion for African Languages

This paper examines the syntactic and semantic expression of basic directional motion and its manner counterparts across the four language phyla of Africa. We sample languages from each phylum, concentrating on basic, non-deictic, directional motion and its counterpart in which motion, direction, and manner of motion occur simultaneously. Two primary findings emerge. Across Africa, basic directional motion tends to be verb-framed with respect to direction; it conflates the semantic components Motion+Direction in verbs equivalent to 'enter, circle, traverse'. Manner of motion, while maintaining this lexicalization pattern, syntactically registers a Motion+Direction verb and its linear order relative to a Manner verb through either of two primary strategies. Manner precedes Motion+ Direction when the two verbs are conveyed through serial verb sequences, verb compounding, or verb coordination. Manner follows Motion+Direction when the former is specified as a deranked verbal form, e.g., as a gerundive, or as part of a deranked clause, e.g., a subordinate clause.

Microstates in Language-Related Brain Potential Maps Show Noun–Verb Differences

Brain processing of grammatical word class was studied analyzing event-related potential (ERP) brain fields. Normal subjects observed a randomized sequence of single German nouns and verbs on a computer screen, while 20-channel ERP field map series were recorded separately for both word classes. Spatial microstate analysis was applied, based on the observation that series of ERP maps consist of epochs of quasi-stable map landscapes and based on the rationale that different map landscapes must have been generated by different neural generators and thus suggest different brain functions. Space-oriented segmentation of the mean map series identified nine successive, different functional microstates, i.e., steps of brain information processing characterized by quasi-stable map landscapes. In the microstate from 116 to 172 msec, noun-related maps differed significantly from verb-related maps along the left–right axis. The results indicate that different neural populations represent different grammatical word classes in language processing, in agreement with clinical observations. This word class differentiation as revealed by the spatial–temporal organization of neural activity occurred at a time after word input compatible with speed of reading.

Deictic explicitness and event continuity in Swahili discourse

This paper addresses the question whether a sequence of two verb forms is to be analyzed as a discontinuous morphological unit (i.e. a compound tense) or a sequence of separate verbs. In the case of Swahili verb sequences consisting of an inflected form of kuwa ‘be’ followed by another inflected verb, syntactic arguments can be advanced for both analyses. Co-occurrence restrictions on permissible tense combinations support a compound tense analysis; the fact that kuwa and V 2 can each be independently negated and can have different subject prefixes supports a two-verb analysis. Discourse data are used to argue in favor of a two-verb analysis. By means of examples from texts and quantitative data it is argued that kuwa supplies deictic orientation for verbs whose orientation is not obvious from context. Such explicit orientation is needed in situations where there is a break in continuity between events, such as introduction of a new subject, that might cause difficulty in integrating a verb into its context. Restrictions on permissible tense combinations in kuwa + V sequences are given a semantic/pragmatic explanation. Implications of the analysis for other languages with relative tense are discussed.

Learning syntax by automata induction

In this paper we propose an explicit computer model for learning natural language syntax based on Angluin's (1982) efficient induction algorithms, using a complete corpus of grammatical example sentences. We use these results to show how inductive inference methods may be applied to learn substantial, coherent subparts of at least one natural language — English — that are not susceptible to the kinds of learning envisioned in linguistic theory. As two concrete case studies, we show how to learn English auxiliary verb sequences (such as could be taking, will have been taking) and the sequences of articles and adjectives that appear before noun phrases (such as the very old big deer). Both systems can be acquired in a computationally feasible amount of time using either positive examples, or, in an incremental mode, with implicit negative examples (examples outside a finite corpus are considered to be negative examples). As far as we know, this is the first computer procedure that learns a full-scale range of noun subclasses and noun phrase structure. The generalizations and the time required for acquisition match our knowledge of child language acquisition for these two cases. More importantly, these results show that just where linguistic theories admit to highly irregular subportions, we can apply efficient automata-theoretic learning algorithms. Since the algorithm works only for fragments of language syntax, we do not believe that it suffices for all of language acquisition. Rather, we would claim that language acquisition is nonuniform and susceptible to a variety of acquisition strategies; this algorithm may be one these.

Learning Syntax by Automata Induction

In this paper we propose an explicit computer model for learning natural language syntax based on Angluin's (1982) efficient induction algorithms, using a complete corpus of grammatical example sentences. We use these results to show how inductive inference methods may be applied to learn substantial, coherent subparts of at least one natural language – English – that are not susceptible to the kinds of learning envisioned in linguistic theory. As two concrete case studies, we show how to learn English auxiliary verb sequences (such as could be taking, will have been taking) and the sequences of articles and adjectives that appear before noun phrases (such as the very old big deer). Both systems can be acquired in a computationally feasible amount of time using either positive examples, or, in an incremental mode, with implicit negative examples (examples outside a finite corpus are considered to be negative examples). As far as we know, this is the first computer procedure that learns a full-scale range of noun subclasses and noun phrase structure. The generalizations and the time required for acquisition match our knowledge of child language acquisition for these two cases. More importantly, these results show that just where linguistic theories admit to highly irregular subportions, we can apply efficient automata-theoretic learning algorithms. Since the algorithm works only for fragments of language syntax, we do not believe that it suffices for all of language acquisition. Rather, we would claim that language acquisition is nonuniform and susceptible to a variety of acquisition strategies; this algorithm may be one these.

The Action-Result (Factitive-Passive) Sequence of Identical Verbs in Biblical Hebrew and Ugaritic

The Action-Result (Factitive-Passive) Sequence of Identical Verbs in Biblical Hebrew and Ugaritic