Verb Similarity Research Articles

Context Sensitive Verb Similarity Dataset for Legal Information Extraction

Existing literature demonstrates that verbs are pivotal in legal information extraction tasks due to their semantic and argumentative properties. However, granting computers the ability to interpret the meaning of a verb and its semantic properties in relation to a given context can be considered as a challenging task, mainly due to the polysemic and domain specific behaviours of verbs. Therefore, developing mechanisms to identify behaviors of verbs and evaluate how artificial models detect the domain specific and polysemic behaviours of verbs can be considered as tasks with significant importance. In this regard, a comprehensive dataset that can be used as an evaluation resource, as well as a training data set, can be considered as a major requirement. In this paper, we introduce LeCoVe, which is a verb similarity dataset intended towards facilitating the process of identifying verbs with similar meanings in a legal domain specific context. Using the dataset, we evaluated both domain specific and domain generic embedding models, which were developed using state-of-the-art word representation and language modelling techniques. As a part of the experiments carried out using the announced dataset, Sense2Vec and BERT models were trained using a corpus of legal opinion texts in order to capture domain specific behaviours. In addition to LeCoVe, we demonstrate that a neural network model, which was developed by combining semantic, syntactic, and contextual features that can be obtained from the outputs of embedding models, can perform comparatively well, even in a low resource scenario.

Measuring of similarity between verb pairs in the biomedical domain

Finding the similarity Ontology-based between two verbs in bio-medical ontology is a difficult task as there is no standard dataset available. This paper is focused on verb-based similarity. So, the similarity between two nouns is taken as the benchmark for working on the verb similarity. There is no exact idea that the verb hierarchy of wordnet is capable to calculate the verb similarity between two verbs. The finding of similarity considers three parameters such as path, link, and depth. But in this paper, in addition to the path, link, and depth parameters, we also considered parameters such as stem-similarity weighting, derivation nouns weighting, and gloss similarity weighting. Moreover, we implemented two algorithms namely Rich Hierarchy Exploration and Shallow Hierarchy Exploration on a dataset and found that there is no significant difference in its performance.

Exploring intrinsic information content models for addressing the issues of traditional semantic measures to evaluate verb similarity

Semantic similarity measures play an important role in many natural language processing and information retrieval activities. It is highly challenging to measure semantic similarity with higher accuracy. A notable branch of semantic similarity evaluation based on information content (IC) is popular in this aspect. Intrinsic information content (IIC) models are another wing of IC based evaluation. Both IC based and IIC based approaches majorly handled similarity evaluation of nouns. Research related to semantic similarity assessment of verb pairs are rarely discussed. To bridge this gap, this work examines various IC based, IIC based approaches on verb pairs. A detailed discussion of the existing measures and their drawbacks are mentioned in this work. Strategies based on information content, length and depth of the concepts are discussed and tested on benchmark datasets. Existing intrinsic information content models are enhanced by addressing various issues like (a) dealing concepts with no path in WordNet and (b) handling the synonym sets of verb concepts. Measures based on path length, intrinsic information content, combined strategies and non-linear strategies for verb pairs are thoroughly inspected. This paper also presents novel strategies to understand novel aspects that are not addressed before. The strategies are experimented by generating the synonym sets of required parts-of-speech which proved very effective in improving the correlation with human judgment. Results on benchmark datasets specify that the proposed approaches for verb similarity will be a guiding factor for understanding the natural language processing tasks.

A method for measuring verb similarity for two closely related languages with application to Zulu and Xhosa


 
 
 There are limited computational resources for Nguni languages and when improving availability for one of the languages, bootstrapping from a related language’s resources may be a cost-saving approach. This requires the ability to quantify similarity between any two closely related languages so as to make informed decisions, of which it is unclear how to measure it. We devised a method for quantifying similarity by adapting four extant similar measures, and present a method of quantifying the ratio of verbs that would need phonological conditioning due to consecutive vowels. The verbs selected are those relevant for weather forecasts for Xhosa and Zulu and newly specified as computational grammar rules. The 52 Xhosa and 49 Zulu rules share 42 rules, supporting informal impressions of their similarity. The morphosyntactic similarity reached 59.5% overall on the adapted Driver-Kroeber metric, with past tense rules only at 99.5%. This similarity score is a result of the variation in terminals mainly for the prefix of the verb.
 
 

A neural classification method for supporting the creation of BioVerbNet

BackgroundVerbNet, an extensive computational verb lexicon for English, has proved useful for supporting a wide range of Natural Language Processing tasks requiring information about the behaviour and meaning of verbs. Biomedical text processing and mining could benefit from a similar resource. We take the first step towards the development of BioVerbNet: A VerbNet specifically aimed at describing verbs in the area of biomedicine. Because VerbNet-style classification is extremely time consuming, we start from a small manual classification of biomedical verbs and apply a state-of-the-art neural representation model, specifically developed for class-based optimization, to expand the classification with new verbs, using all the PubMed abstracts and the full articles in the PubMed Central Open Access subset as data.ResultsDirect evaluation of the resulting classification against BioSimVerb (verb similarity judgement data in biomedicine) shows promising results when representation learning is performed using verb class-based contexts. Human validation by linguists and biologists reveals that the automatically expanded classification is highly accurate. Including novel, valid member verbs and classes, our method can be used to facilitate cost-effective development of BioVerbNet.ConclusionThis work constitutes the first effort on applying a state-of-the-art architecture for neural representation learning to biomedical verb classification. While we discuss future optimization of the method, our promising results suggest that the automatic classification released with this article can be used to readily support application tasks in biomedicine.

Similitud verbal: Análisis comparativo entre lingüística teórica y datos extraídos de corpus

The concept of similarity is crucial for fields belonging to cognitive science and linguistics. However, it is a broad notion that can be approached through multiple perspectives. In this work we carry out a comparative analysis on how two of these perspectives articulate similarity between verb senses in Spanish. The first perspective is based on corpus linguistics, specifically on semantic roles annotated on a Spanish corpus, as representation of argument structure associated to verb senses. The second perspective belongs to the field of theoretical linguistics and is embodied by the constructions that a verb instantiates. With such comparison we aim to investigate whether argument structure obtained from corpus, which reflects the expression of an event, configures verb similarity in a manner congruent with the way in which constructions do it, which are argued to contain argument structure too and to constitute key elements in the generation of sentence meaning. The results of the comparison show that there is a significant correlation between the similarity formalization of both perspectives, particularly when we take into account syntactic and pattern information from corpus. Additionally, the results attribute particular importance to the aspectual opposition of stativity-dynamicity as the backbone of the similarity-dissimilarity axis for constructions.

Verb Forms Functioning in Cyril Turovskij’s Homilies

Summary The article issue is verbs functioning in the works of Old Russian writer Cyril Turovskij (the second half of the 12th century). It is being considered in details for the first time on the base of the earliest source, Tolstovskij Sbornik (the second half of the 13th century). Since Cyril’s homilies were addressed to a wide range of listeners and readers, they had to be based on clear and straightforward language forms related, however, to literary standards. The article describes following features of Cyril’s language: the earliest usage of “praesens historicum”; exclusive usage of aorist forms with additional endings (начатъ type); distinctive character of aorist рѣша; unusually broad usage of 2 Sg. aorist forms; usage of perfective imperfect forms and imperfects with additional endings; standardized usage of Future I forms; individual cases of imperative usage. A certain similarity of verbs functioning in Cyril’s homilies and The Tale of Igor’s Campaign is detected.

Children's Acquisition of the English Past-Tense: Evidence for a Single-Route Account From Novel Verb Production Data.

This study adjudicates between two opposing accounts of morphological productivity, using English past‐tense as its test case. The single‐route model (e.g., Bybee & Moder, 1983) posits that both regular and irregular past‐tense forms are generated by analogy across stored exemplars in associative memory. In contrast, the dual‐route model (e.g., Prasada & Pinker, 1993) posits that regular inflection requires use of a formal “add ‐ed” rule that does not require analogy across regular past‐tense forms. Children (aged 3–4; 5–6; 6–7; 9–10) saw animations of an animal performing a novel action described with a novel verb (e.g., gezz; chake). Past‐tense forms of novel verbs were elicited by prompting the child to describe what the animal “did yesterday.” Collapsing across age group (since no interaction was observed), the likelihood of a verb being produced in regular past‐tense form (e.g., gezzed; chaked) was positively associated with the verb's similarity to existing regular verbs, consistent with the single‐route model only. Results indicate that children's acquisition of the English past‐tense is best explained by a single‐route analogical mechanism that does not incorporate a role for formal rules.

Verb similarity: Comparing corpus and psycholinguistic data

Abstract Similarity, which plays a key role in fields like cognitive science, psycholinguistics and natural language processing, is a broad and multifaceted concept. In this work we analyse how two approaches that belong to different perspectives, the corpus view and the psycholinguistic view, articulate similarity between verb senses in Spanish. Specifically, we compare the similarity between verb senses based on their argument structure, which is captured through semantic roles, with their similarity defined by word associations. We address the question of whether verb argument structure, which reflects the expression of the events, and word associations, which are related to the speakers’ organization of the mental lexicon, shape similarity between verbs in a congruent manner, a topic which has not been explored previously. While we find significant correlations between verb sense similarities obtained from these two approaches, our findings also highlight some discrepancies between them and the importance of the degree of abstraction of the corpus annotation and psycholinguistic representations.

The Distribution of Words in Chinese and Laos Based on Cross Language Corpus

Word representation is the basic research content of natural language processing. At present, distributed representation of monolingual words has shown satisfactory application effect in some Neural Probabilistic Language (NPL) research, while as for distributed representation of cross-lingual words, there is little research both at home and abroad. Aiming at this problem given distribution similarity of nouns and verbs in these two languages, we embed mutual translated words, synonyms, super-ordinates into Chinese corpus by the weakly supervised learning extension approach and other methods, thus Laos word distribution in cross-lingual environment of Chinese and Laos is learned. We applied the distributed representation of the cross-lingual words learned before to compute similarities of bilingual texts and classify the mixed text corpus of Chinese and Laos, Experimental results show that the proposal has a satisfactory effect on the two tasks.

Old Slavic Sermon Language: The Extraordinary Nature of Verb Morphology in Cyril Turovskij’s Homilies

The article’s subject matter—verbs functioning in the sermons of the Old Russian church writer Cyril Turovskij (second half of the 12th century)—is considered in details for the first time on the basis of the earliest source, Tolstovskij Sbornik (second half of the 13th century). Since Cyril’s sermons were addressed to a wide range of listeners and readers they had to be based on intelligible and simple language forms that also preserved a connection with literary standards. This manifested itself in the significant Russification of the preaching language. The article describes the following features of the language of Cyril’s sermons: the earliest and widespread usage of “praesens historicum”; the exclusive usage of aorist forms with additional endings (načętъ type); the special functional and syntactic nature of the aorist rěšę; the unusually wide usage of 2 Sg. aorist and imperfect forms; the usage of perfective imperfect forms and imperfects with additional endings; the prospective future tense and modal functioning of the paraphrastic forms with the auxiliary verb xoščè; special cases of 1 Pl. imperatives usage; the special character of the reflexive enclitic sę; and the extraordinary distribution of periphrastic preterits forms. Some similarity of verbs functioning in Cyril’s homilies and The Tale of Igor’s Campaign is detected as well as in the original Chronicle, early Old Russian translations, and Paroemiarion.

Suggesting Alternative Scenarios Using Use Case Specification Patterns for Requirement Completeness

Completeness in software requirements specification is one of the key factors for successful software development. For specifying software requirements, scenario-based approach is broadly used, comprising a basic flow regarding the successful use of the system, and alternative flows describing abnormal or less frequent interactions of the system. However, alternative flows tend to be frequently missed in many cases, because of the relative lower significance rather than the basic flow, which eventually have an influence on achieving the completeness of software requirements. In order to address the issue, we propose an approach for automatically recommending alternative flows from a basic flow by extracting the essential use case patterns based on the occurrence patterns of the agents and measuring the verb similarity between the main verbs of each scenario. In order to validate our approach, we apply it to three industrial case studies, and show comprehensiveness of the suggested alternative flows and synergic effectiveness for inexperienced developers.

The perfective past tense in Greek adolescents with Down Syndrome

This study investigates the ability of a group of eight Greek-speaking adolescents with Down Syndrome (DS) (aged 12.1–18.7) to handle the perfective past tense using an acceptability judgement task. The performance of the DS participants was compared with that of 16 typically-developing children whose chronological age was matched with the mental age of the DS group. For existing verbs, both groups showed high accuracy scores for the sigmatic past tense whilst for (potential but non-existing) nonce verbs the DS group performed differently from the controls. Specifically, their judgements were unaffected by a nonce verb's similarity to existing verbs, unlike those of the controls, suggesting that the DS participants were less reliant on similarity-based generalisations when encountering a nonce word than the controls. Apart from that, it was found that people with DS did not show any kind of morphological impairment, replicating previous findings on past tense production in DS.

FrameNet's Frames vs. Levin's Verb Classes

The classification of verbs in Levin's (1993) English Verb Classes and Alternations: A preliminary Investigation, on the basis of both intuitive semantic grouping and their participation in valence alternations, is often used by the NLP community as evidence of the semantic similarity of verbs (Jing & McKeown 1998; Lapata & Brew 1999; Kohl et al. 1998). In this paper, we compare the Levin classification with the work of the FrameNet project (Fillmore & Baker 2001), where words (not just verbs) are grouped according to the conceptual structures (frames) that underlie them and their combinatorial patterns are inductively derived from corpus evidence. This means that verbs grouped together in FrameNet (FN) might be semantically similar but have different (or no) alternations, and that verbs which share the same alternation might be represented in two different semantic frames.

Analogy and the dual-route model of morphology

Abstract Prasada and Pinker's (1993) subjects provided past tense forms of nonce verbs. The subject's willingness to provide irregular past tense forms correlated with the verb's phonological similarity to existing irregular English verbs. However, there was no correlation between the number of nonce verbs assigned regular inflection, and the verb's similarity to existing regular verbs. According to the dual-route model, this is expected since irregular items are stored in associative memory, while regular items take an allomorph of -ed by rule. A singleroute connectionist simulation failed to duplicate the subject's behavior on regular verbs. Two instance-based models were applied to the data: Analogical Modeling of Language and the Tilburg Memory Based Learner. Each model employed a similarity algorithm to determine the behavior of all regular and irregular items. Both models successfully mirrored the subject's responses. Therefore, the data are consistent with an instance-based single-route model of morphology.

Concepts of Mental Activities and Verbs in Children of High and Average Verbal Intelligence

In order to understand the contribution of increased cognitive abilities to children's naive theories of mind, children of high and average verbal intelligence rated their understanding of the interrelationships between and among mental activities and verbs. In Study 1, participants rated the similarity of pairs of prototypical mental activity scenarios according to the "way you use your mind" in each one. The scenarios represented the categories of list memory, prospective memory, comprehension, selective attention, inference, planning, comparison, and recognition. In Study 2, participants rated the similarity of cognitive and affective verbs (e.g., decide, memorize, love, worry). Multidimensional scaling and clustering analyses indicated very similar organization and structure of concepts in children of high and average verbal intelligence in both studies. These results add to the growing body of literature suggesting that metacognitive development and its association with intelligence differ depending on the type of metacognition being examined (Alexander, Carr, & Schwanenflugel, 1995).

Theory of Mind in Deaf Adults and the Organization of Verbs of Knowing

Naive theories of mind provide an organizing scheme for concept formation and categorization. Additionally, they highlight what is important within a domain. This study investigated how deaf adults with hearing parents orgaize 17 cognitive verbs of knowing as a way of describing their naive theory of mind. Deaf adults rated on 1 to 7 scale the similarity of pairs of cognitive verbs in terms of whether 'the words are alike or different based on how you would use your mind when you do that mental activity'. We directly compared the similarity of cognitive verbs in these deaf adults with data collected in earlier research describing the organisation of cognitive verbs in hearing adults. We conducted multidimensional scaling, additive similarity tree, and Pathfinder analyses to assess global, categorical, and local relations in the domain. Deaf adults' theory of mind revealed distinction among mental verbs in terms of information-processing components and constructive certainty components. In all analyses, the deaf group showed a very similar organization to that of hearing adults examined in previous research. We conclude that, although deaf adults might be expected to view congnitive processes differently than hearing adults, they nonetheless exhibit a theory of mind that is highly similar to that of hearing adults.

Recognition memory for elements of sentences

Although several theoretical positions and a variety of empirical tasks indicate the importance of verbs to sentences, nouns are generally recalled and recognized better in memorial tasks. Three main models can be identified to explain this discrepancy ("Fillenbaum's paradox"). To try to resolve this paradox, several experiments explored the efficiency of various sentence elements as cues in recognition memory. In Experiment I, concreteness of the stimuli did not interact with the type of distractor; however, verb phrase changes were harder to recognize than noun phrase changes when synonym distractors were used. This result was replicated in a forced-choice recognition paradigm (Experiment II) and with whole sentences where the derivational similarity of verbs and nouns was controlled (Experiment IV). The effect could not be attributed to characteristics of the English language (Experiment III) or to superior memory for form information in nouns (Experiment V). The total results are interpreted as suggesting that subjects process different parts of a sentence to different semantic levels, with verbs receiving more semantic representation and nouns more orthographic or phonological representation. The results are taken as support for a "semantic encoding model" of Fillenbaum's paradox.