Prepositional Phrase Attachment Research Articles

An investigation of the lexico-grammatical profile of English legal- lay language

The article presents a study on the lexico-grammar of the genre ofEnglish legal-lay language (Tiersma 1999), using the English subcorpus of theCorIELLS corpus (Busso forthcoming). The study explores four grammaticalconstructions (in Goldberg 2006’s Construction Grammar sense): nominalisationsheading prepositional phrase attachments, modal verb constructions, participialreduced relative constructions, and passive constructions. Specifically, we usecollostructional analysis (Stefanowitsch 2013), followed by a vocabulary analysisusing English core vocabulary as a reference (Brezina and Gablasova 2015), anda comparative frequency analysis with corpora of legal language and general-domain written prose. Results of this first part of the study foreground how legal-lay language is quantitatively different from both neighbouring genres, suggestingthat it might be considered a “blended” genre. We further explore the data in termsof accessibility for speakers, using readability metrics and a survey on Englishparticipants. Both methods show that legal-lay language is at an intermediatelevel of complexity between legal jargon and general-domain prose; however, wefurther note that readability metrics generally underestimate speakers’ ability tocomprehend legal-lay language.

Relative Clause Effects at the Matrix Verb Depend on Type of Intervening Material.

Although a large literature demonstrates that object-extracted relative clauses (ORCs) are harder to process than subject-extracted relative clauses (SRCs), there is less agreement regarding where during processing this difficulty emerges, as well as how best to explain these effects. An eye-tracking study by Staub, Dillon, and Clifton (2017) demonstrated that readers experience more processing difficulty at the matrix verb for ORCs than for SRCs when the matrix verb immediately follows the relative clause (RC), but the difficulty is eliminated if a prepositional phrase (PP) intervenes. A careful examination of Staub etal.'s materials reveals that the types of PPs used in the experiment were a mixture of locative and temporal PPs. This is important in that locative PPs can modify either a noun phrase or a verb phrase (VP), whereas temporal PPs typically modify VPs, resulting in systematic differences in PP attachment across ORCs versus SRCs. In the current eye-tracking experiment, we systematically manipulated RC type and PP type in the same sentences used by Staub etal. The manipulation of PP type resulted in a crossover pattern at the matrix verb such that there was a trend for reading times to be longer for ORCs than SRCs when the PP was locative, but reading times were longer for SRCs than ORCs when the PP was temporal. These results provide important information regarding the locus of RC-processing effects and highlight the importance of carefully considering how intervening material might unintentionally alter the structure or the meaning of a sentence.

A Theoretical Foundation for Syntactico-Semantic Pattern Recognition

Conventionally syntactic pattern recognition tasks have been driven by grammars defining a syntactic structure. Syntactic Pattern recognition tasks were primarily relying on the ability of parsing algorithms to recognize the patterns in the input data. These algorithms essentially were dependent on the syntactic grammars defining the patterns. Context free grammars, a type of grammars have been particularly well studied for pattern recognition tasks to be solved by computer efficiently. Some of the key pattern recognition tasks had applications in Natural Language Processing (NLP). Though context free grammars are well suited for capturing rigid patterns and unambiguous patterns, there was a need to encapsulate the uncertainty aspects involved in some pattern recognition processes. Probabilistic context free grammars can well handle the need to capture uncertainty in the processes but not in a true sense they are able to capture the uncertainty associated with the semantic context governing the domain in which the pattern recognition processes are being attempted at. The paper formally puts forth an approach for syntactico-semantic pattern recognition. The syntactico-semantic pattern recognition attempts to capture the semantic context and the uncertainties involved thereof along with probabilistic reasoning. The approach consists of integration mapping between probabilistic context free grammar (PCFG) and Multi Entity Bayesian network (MEBN), a first-order logic for modeling probabilistic knowledge bases. Additionally, the paper outlines a modified version of the CYK parser algorithm for the defined mapping between PCFG and MEBN with a method to ensure the properness and consistency of such PCFG along with its key application, disambiguation of PP (Prepositional Phrase) attachment. The theoretical foundation proposed has been validated by a proof-of-concept implementation of the modified CYK algorithm for syntactico-semantic reasoning in Java with promising ability to disambiguate PP attachment uses cases of New York Times and Wikipedia corpus dataset samples.

The Effect of Ambiguity Awareness on Second Language Learners' Prosodic Disambiguation.

Three tasks were reported to examine the effect of ambiguity awareness on Chinese-speaking English learners’ use of prosody in resolving prepositional-phrase attachment ambiguity. In the first (Task 1) and second (Task 2) tasks, listeners were not informed of the syntactic ambiguity. In the third task (Task 3), listeners were given the specific information about syntactic ambiguity. The analysis of the overall accuracy rate showed that before receiving specific information about syntactic ambiguity, learners did not detect the ambiguity within the structure and tended to interpret the sentence in a “good-enough” heuristic to reduce the computational burden. After being aware of the syntactic ambiguity, they could use prosodic cues to resolve the ambiguity. However, the finding that the learners reversed their parsing bias from verb phrase attachment (VP-attachment) toward noun phrase attachment (NP-attachment) indicated their difficulty in integrating prosodic information to syntactic structure efficiently. The analysis of individual accuracy rate demonstrated learners’ individual variations in using prosodic cues. The result suggests that learners’ failure to use prosodic cues may be attributed to a lack of ambiguity awareness and difficulty in information integration, rather than their low sensitivity to prosodic cues.

Multimodal Machine Learning for Natural Language Processing: Disambiguating Prepositional Phrase Attachments with Images

Although documents are increasingly multimodal, their automatic processing is often monomodal. In particular, natural language processing tasks are typically performed based on the textual modality only. This work extends the syntactic parsing task to the image modality in addition to text. In particular, we address the prepositional phrase attachment problem, a hard and semantic problem for syntactic parsers. Given an image and a caption, the proposed approach resolves syntactic attachment of prepositions in the parse tree according to both visual and lexical features. Visual features are derived from the nature and position of detected objects in the image that are aligned to textual phrases in the caption. A reranker uses this information to reorder syntactic trees produced by a shift-reduce syntactic parser. Trained on the Flickr-PP corpus which contains multimodal gold-standard attachments, this approach yields improvements over a text-only syntactic parser, in particular for the subset of prepositions that encode location, leading to an increase of up to 17 points of attachment accuracy.

Using Word Sketches to Resolve Prepositional Phrase Attachment Ambiguity in Arabic

Using Word Sketches to Resolve Prepositional Phrase Attachment Ambiguity in Arabic

Predicting the Argumenthood of English Prepositional Phrases

Distinguishing between arguments and adjuncts of a verb is a longstanding, nontrivial problem. In natural language processing, argumenthood information is important in tasks such as semantic role labeling (SRL) and prepositional phrase (PP) attachment disambiguation. In theoretical linguistics, many diagnostic tests for argumenthood exist but they often yield conflicting and potentially gradient results. This is especially the case for syntactically oblique items such as PPs. We propose two PP argumenthood prediction tasks branching from these two motivations: (1) binary argumentadjunct classification of PPs in VerbNet, and (2) gradient argumenthood prediction using human judgments as gold standard, and report results from prediction models that use pretrained word embeddings and other linguistically informed features. Our best results on each task are (1) acc. = 0.955, F1 = 0.954 (ELMo+BiLSTM) and (2) Pearson’s r = 0.624 (word2vec+MLP). Furthermore, we demonstrate the utility of argumenthood prediction in improving sentence representations via performance gains on SRL when a sentence encoder is pretrained with our tasks.

When actions and looks don’t line up

Abstract In two eye tracking experiments, we investigate how adult child-L2 speakers of English resolve prepositional phrase (PP) attachment ambiguity in their dominant language (English), and whether they use prosodic information to aid in the process of garden-path recovery. The findings showed an increased processing cost associated with the revision of temporary ambiguous sentences for the child-L2 adults relative to the native English speakers. When prosody was informative, the child-L2 adults were able to use prosodic information to guide the interpretation of their later acquired, dominant language. However, they performed revision significantly less successfully than the native speakers. Although processing was similar for the native English speakers and the adult child-L2 speakers of English, when it comes to sensitivity to prosodic information and referential context, the two groups differed with regards to reanalysis both in the presence and absence of salient prosodic and referential information.

Prepositional Phrase Attachment Global Ambiguity Resolution in Semantically Biased and Neutral Conditions by L2 Learners

This study investigated the prepositional phrase attachment preferences of Persian speaking second language learnersof English in dealing with globally ambiguous sentences. Results are reported from 34 participants across twoproficiency levels who took part in an online reading task and an offline reading task. In both tasks, the preferenceswere examined in biased and neutral conditions; in the former the verb was semantically biased towards the DPinside the PP (e.g. Peter killed the man with a gun) and in the latter condition there was no bias (e.g. Peter met theman with a gun). The findings showed that in both online and offline reading tasks participants resolved theambiguity by attaching the ambiguous DP to the verb, being influenced by the semantic relationship between the two,but in the neutral condition, the preferences were almost equal for both interpretations. The proficiency level did notinfluence preferences except for a slight interaction effect in the offline reading task. The findings indicate that L2learners of different proficiency levels rely on semantics rather than syntax in processing globally ambiguoussentences. The findings of this study in the neutral condition, are in line with the unrestricted race model. However,this model and the constraint-based theories fail to provide a proper explanation for the biased and neutral conditionsrespectively. It is suggested that discrepancy between the assumptions of the constraint-based theories and theunrestricted race model seems to be due to ignoring the role of semantic relationship between phrases in sentencesused for elicitation of parsing preferences; a factor that needs to be taken into account in future studies.

Prosodic Boundary Effects on Syntactic Disambiguation in Children With Cochlear Implants.

This study investigated prosodic boundary effects on the comprehension of attachment ambiguities in children with cochlear implants (CIs) and normal hearing (NH) and tested the absolute boundary hypothesis and the relative boundary hypothesis. Processing speed was also investigated. Fifteen children with NH and 13 children with CIs (ages 8-12 years) who are monolingual speakers of Brazilian Portuguese participated in a computerized comprehension task with sentences containing prepositional phrase attachment ambiguity and manipulations of prosodic boundaries. Children with NH and children with CIs differed in how they used prosodic forms to disambiguate sentences. Children in both groups provided responses consistent with half of the predictions of the relative boundary hypothesis. The absolute boundary hypothesis did not characterize the syntactic disambiguation of children with CIs. Processing speed was similar in both groups. Children with CIs do not use prosodic information to disambiguate sentences or to facilitate comprehension of unambiguous sentences similarly to children with NH. The results suggest that cross-linguistic differences may interact with syntactic disambiguation. Prosodic contrasts that affect sentence comprehension need to be addressed directly in intervention with children with CIs.

Syntactic Ambiguity Resolution of Cantonese-English Bilinguals: Evidence from Prepositional Phrase and Adverb

Syntactic ambiguity resolution is influenced by multiple constraints. A sentence-picture matching task tested the attachment of prepositional phrase (PP) such as on the chair within an utterance like Mary ate the apple on the chair. PP attaching to NP2 (the apple) is called low attachment (LA), while PP attaching to V (ate) is called high attachment (HA). A comprehension task tested the attachment of adverb such as quickly within an utterance like the story Mary was reading quickly put her little sister into sleep. The checked adverbs attaching to the verb of the embedded clause (VEC) and to the verb of the main clause (VMC) were analyzed. Cantonese-English bilinguals were grouped according to their L2 proficiency. Results reflected that 1) both groups showed preference to HA for PP, 2) LP group showed preference to VMC for adverb attachment; 3) there is a large gap in LA between LP and HP, suggesting that syntactic ambiguity resolution involves interaction of all the information available. The relationship between language proficiency and parsing is not so definitely supported by the results.

An Analysis of Prepositional-Phrase Attachment Disambiguation

An Analysis of Prepositional-Phrase Attachment Disambiguation

Resolving vision and language ambiguities together: Joint segmentation & prepositional attachment resolution in captioned scenes

We present an approach to simultaneously perform semantic segmentation and prepositional phrase attachment resolution for captioned images. Some ambiguities in language cannot be resolved without simultaneously reasoning about an associated image. If we consider the sentence “I shot an elephant in my pajamas”, looking at language alone (and not using common sense), it is unclear if it is the person or the elephant wearing the pajamas or both. Our approach produces a diverse set of plausible hypotheses for both semantic segmentation and prepositional phrase attachment resolution that are then jointly re-ranked to select the most consistent pair. We show that our semantic segmentation and prepositional phrase attachment resolution modules have complementary strengths, and that joint reasoning produces more accurate results than any module operating in isolation. Multiple hypotheses are also shown to be crucial to improved multiple-module reasoning. Our vision and language approach significantly outperforms the Stanford Parser (De Marneffe et al., 2006) by 17.91% (28.69% relative) and 12.83% (25.28% relative) in two different experiments. We also make small improvements over DeepLab-CRF (Chen et al., 2015).

Accents, Not Just Prosodic Boundaries, Influence Syntactic Attachment.

Traditionally, pitch accents are understood to relate to the information structure of a sentence and its discourse connections, while prosodic boundaries indicate groupings of words and affect how constituents attach into a syntactic structure. Here, we show that accents also affect syntactic attachment in multiple different syntactic structures. Three auditory questionnaires on ambiguous attachment sentences (such as Tom reported that Bill was bribed [last May]) find that accenting the higher or lower verb ( reported or bribed) increases the attachment of the final adverbial phrase as a modifier of the accented verb. A fourth experiment shows that accents on verbs or object nouns (in sentences like Jenny sketched a child [with crayons]) also increase the attachment of the final prepositional phrase to the accented head (sketched with crayons versus a child with crayons). Accent effects were small but consistent across sentences with different levels of bias and did not depend on prosodic boundaries. The results suggest that focused elements are important to the main assertion of the sentence and therefore draw the attachment of upcoming material (though the salience of attachment sites may also be important). The results also demonstrate that both prosodic phrasing and pitch accents can affect basic syntactic structure.

Syntactic Ambiguity of Arabic Prepositional Phrase Attachment in Saudi Newspaper Headlines

In this article, an attempt is made to analyze syntactic ambiguity in a specific register, i.e. newspaper headlines. As a form of media text, newspaper has a significant role in society as it presents various local, national and international events and information. In order to attract readers' attention, journalists may tend to form headlines to be ambiguous and confusing so that readers become curious to read the whole report. This article addresses the issue of syntactic ambiguity caused by prepositional phrase (PP) attachment in newspaper headlines by analyzing examples of such headlines. After investigating the ambiguity of the headlines, an attention was directed to PP attachment preferences by examining the type of headlines (VP-NP-PP and NP1-NP2-PP structure). The main results were that NP attachment was generally preferred over VP attachment and NP1 attachment was notably favored over NP2 attachment.

L2 Learners’ English Prepositional-Phrase Processing with Consideration of L2 Proficiency and Word Familiarity: A mixed-effect Analysis

L2 Learners’ English Prepositional-Phrase Processing with Consideration of L2 Proficiency and Word Familiarity: A mixed-effect Analysis Hiroki FUJITA Abstract A considerable number of studies have hitherto examined how native speakers and second language (hereafter L1 and L2, respectively) learners of English process English prepositional phrases. It is usually thought that both L1 and L2 speakers have a preference for verb phrase modification (e.g., The boy broke the old window with a ball of snow.) over noun phrase modification (e.g., The boy broke the old window with a view of snow.). However, little research has attempted to reveal how non-structural factors interact with L2 learners’ prepositional phrase processing. In this study, I examined whether L2 learners’ syntactic processing preference for prepositional-phrase attachment was affected by their L2 proficiency and word familiarity with reading time data obtained from the word-by-word self-paced reading task. In addition, the possibility that L2 syntactic priming is influenced by such non-syntactic factors was also investigated. The research results showed that L2 learners’ prepositional-phrase processing and syntactic priming were not subject to those factors, indicating that in prepositional phrases, there was no interaction between L2 learners’ syntactic processing and those non-syntactic elements. Full Text: PDF DOI: 10.15640/jehd.v5n1a10

Sentence-Identity Building: A Syntactic Resolution Of Amphiboly Caused By Prepositional-Phrase Attachment

Abstract Amphiboly is a matter of linguistic ambiguity, which, in its turn, is essentially related to the intractable natural issue of the disparity between what we can possibly think and what we can actually express through language. The papers on various types of ambiguity and the solutions to them are not few. However, as with almost everything of a theoretical nature in language, they are far from being enough to cover the complexity of the issue in point. The present paper does not claim to solve amphiboly once and for all: it only proposes a framework that can be followed easily by those interested in the interpretation of similar sentences and in the production of unequivocal ones. The material selected for analysis consists in a few sentences, collected mainly from online sources, in which amphiboly caused by prepositional-phrase attachment is present. Some instruments provided by Syntax and Morphology as branches of Linguistics are used to interpret them so that the framework of reasoning that we provide can be used whenever a similar structure needs disambiguation.

한국인 학습자들의 영어 전치사구 부가 모호성 해결 책략에 관한 연구

한국인 학습자들의 영어 전치사구 부가 모호성 해결 책략에 관한 연구

Exploring Compositional Architectures and Word Vector Representations for Prepositional Phrase Attachment

Prepositional phrase (PP) attachment disambiguation is a known challenge in syntactic parsing. The lexical sparsity associated with PP attachments motivates research in word representations that can capture pertinent syntactic and semantic features of the word. One promising solution is to use word vectors induced from large amounts of raw text. However, state-of-the-art systems that employ such representations yield modest gains in PP attachment accuracy. In this paper, we show that word vector representations can yield significant PP attachment performance gains. This is achieved via a non-linear architecture that is discriminatively trained to maximize PP attachment accuracy. The architecture is initialized with word vectors trained from unlabeled data, and relearns those to maximize attachment accuracy. We obtain additional performance gains with alternative representations such as dependency-based word vectors. When tested on both English and Arabic datasets, our method outperforms both a strong SVM classifier and state-of-the-art parsers. For instance, we achieve 82.6% PP attachment accuracy on Arabic, while the Turbo and Charniak self-trained parsers obtain 76.7% and 80.8% respectively.

Priming prepositional phrase attachment: evidence from eye-tracking and event-related potentials.

Three syntactic-priming experiments investigated the effect of structurally similar or dissimilar prime sentences on the processing of target sentences, using eye tracking (Experiment 1) and event-related potentials (ERPs) (Experiments 2 and 3) All three experiments tested readers' response to sentences containing a temporary syntactic ambiguity. The ambiguity occurred because a prepositional phrase modifier (PP-modifier) could attach either to a preceding verb or to a preceding noun. Previous experiments have established that (a) noun-modifying expressions are harder to process than verb-modifying expressions (when test sentences are presented in isolation); and (b) for other kinds of sentences, processing a structurally similar prime sentence can facilitate processing a target sentence. The experiments reported here were designed to determine whether a structurally similar prime could facilitate processing of noun-attached modifiers and whether such facilitation reflected syntactic-structure-building or semantic processes. These findings have implications for accounts of structural priming during online comprehension and for accounts of syntactic representation and processing in comprehension.