Adjective Combinations Research Articles

Evaluation of adjective and adverb types for effective Twitter sentiment classification.

Twitter, the largest microblogging platform, has reported more than 330 million active users in recent years. Many users express their sentiments about politics, sports, products, personalities, etc. Sentiment analysis has emerged as a specialized branch of machine learning in which tweets are binary-classified to provide sentimental insights. A major step in sentiment classification is feature selection, which primarily revolves around parts of speech (POS). Few techniques merely focused on single features such as adjectives, adverbs, and verbs, while other techniques examined types of these features, such as comparative adjectives, superlative adjectives, or general adverbs. Furthermore, POS as linguistic entities have also been studied and extensively classified by researchers, such as CLAWS-C7. For sentiment analysis, none of the studies conceptualized all possible POS features under similar conditions to draw firm conclusion. This research is centered on the following objectives: 1) examining the impact of various types of adjectives and adverbs that have not been previously explored for sentiment classification; 2) analyzing potential combinations of adjectives and adverbs types 3) conducting a comparison with a benchmark dataset for better classification accuracy. To assess the concept, a renowned human annotated dataset of tweets is investigated. Results showed that classification accuracy for adjectives is improved up to 83% based on the general superlative adjective whereas for adverbs, comparative general adverb also depicted significant accuracy improvement. Their combination with general adjectives and general adverbs also played a substantial role. The unexplored potential of adjectives and adverb types proved better in accuracy against state-of-the-art probabilistic model. In comparison to lexicon-based model, proposed research model overruled the dependency of lexicon-based dictionary where each term first needs to be matched for semantic orientation. The evident outcomes also help in time reduction aspect where huge volume of data need to be processed swiftly. This noteworthy contribution brought up significant knowledge and direction for domain experts. In the future, the proposed technique will be explored for other types of textual data across different domains.

COVID vocabulary in the Bulgarian media discourse and in the official state regulatory documentation (in view of innovative trends in Bulgarian grammar and (non)deviations from spelling rules)

The article examines the vocabulary related to the coronavirus pandemic, in view of innovative trends in Bulgarian grammar and in view of (non)deviations from spelling rules. The conclusions were drawn based on the analysis of examples from 38 regulatory documents (orders, instructions, reports), 107 excerpted word combinations of invariable adjectives and nouns, and 20 other language units from the media discourse. The emphasis of the study is on the analysis of the semantic-structural and spelling features of the appositive word combinations of invariable COVID adjectives and variable nouns. Other COVID names related to the appearance of new biaspectual verbs, verbal nouns and the high frequency (frequency of use) of substantivized participles and nouns are also analyzed.

A Cognitive Approach to Expressions of English Metaphors

This paper analyses the three expressions of English metaphors: nominal metaphors; verbal metaphors; adjective metaphors from the perspective of cognitive linguistics. For nominal metaphors, the author starts from metaphorical mapping mechanism, interpreting the different syntactic features of them: as a subject, an object, a predicate, an appositive and a possessive, figuring out the reasons for producing nominal metaphors and its cognitive approach. For verbal metaphors, this paper has, in detail, analyzed three different categories of them: the verbal metaphors with a logical subject predicate collocation; the ones with a logical verb, object collocation; and with a logical subject, predicate, object collocation with the above three kinds of linguistic theories, obtaining the forming mechanism of the verbal metaphors and its working ways in languages. For the adjective metaphors, this paper mainly applies the mapping theory between a source domain and a target domain in Conceptual Integration Theory. A comprehensive, detailed and systematical analysis of adjective meanings in basic adjective metaphors and metaphorical combinations of basic adjectives leads to the following conclusion: the combination of basic adjective metaphors is the result of the “Least Effort Principle” with characteristics of simplicity and popularity. In addition, it also possesses the capability to produce richer metaphorical expressions.

Chinese Two-Character Word Orthographic Processing and Adjective-Noun Collocation Comprehension in Sentence Reading.

Reading can be regarded as a combination of lexical decoding and linguistic comprehension (Hoover and Gough in Read Writ Interdiscip J 2:127-160, 1990). In Chinese sentence reading, skilled readers' difficulties in phonological processing significantly enhance the 'wrap-up' effect (Li and Lin in J Deaf Stud Deaf Educ 25(4):505-516, 2020). To examine how orthographic processing in Chinese two-character word recognition might interact with adjective-noun collocation (ANC) comprehension before the wrap-up effect, two experiments were conducted in the same paradigm as used by Li and Lin (J Deaf Stud Deaf Educ 25(4):505-516, 2020). The sentences contained ANCs or semantically inappropriate combinations of adjectives and nouns (nANCs), the adjectives (Experiment 1) or the nouns (Experiment 2) of which were two-character words or corresponding transposed nonwords (T-nonwords). Similar results were obtained in both experiments: difficulties in T-nonword processing and in nANC comprehension collectively lengthened the reading times of the words immediately following. In conclusion, sentence reading likely contains interactions between orthographic processing and linguistic comprehension. As an indication of psycholinguistic significance, skilled readers have to use extra resources to suspend the cognitive vigilance that arises from unexpected demand in lexical decoding, in addition to their main focus of linguistic comprehension.

Acehnese lexical and grammatical collocations of the North Aceh dialect

This research dealt with collocations used in the North Aceh dialect. It analyzed the part of speech categories into which collocations of the North Aceh dialect can be grouped. This research focused on the grammatical collocations and lexical collocations used in the Blang Mee village of Bireuen District, Aceh, Indonesia. This is a descriptive qualitative using case study looking into the villagers’ use of Acehnese collocations. Six Acehnese speakers were selected as the language informants. They were fifty years old or above and never traveled or lived outside of Blang Mee. Data were extracted from interviews with these speakers who were asked to discuss general topics in Acehnese. The analysis was performed using a framework on collocation categories by Benson et al. The results of the analysis showed both lexical and grammatical collocations used by speakers in North Aceh. Lexical collocations were found in the forms of verb (denoting creation or activation) + noun combinations, verb (meaning eradication or nullification) + a noun, adjective + noun combinations, noun + verb combinations, noun + noun combinations, adverb + adjective combinations, and verb + adverb combinations. Grammatical collocations came in the following combinations: noun + preposition combinations, adjective + preposition combinations, preposition + noun combinations, and collocational verb patterns. The collocations used by the North Aceh dialect speakers indicate the uniqueness of their dialect within other dialects spoken by the Acehnese.

English adjectives of very similar meaning used in combination: an exploratory, corpus-aided study

This article reports on research into the phenomenon whereby English adjectives of very similar meaning are sometimes used in combination, either as an uncoordinated sequence (as in ‘simple straightforward’) or as a coordinated pair (as in ‘dull and monotonous’). One of the preliminary goals of the research is to identify areas of meaning in which such adjective combinations typically occur. In order to find relevant examples, the British National Corpus was used as a source of textual material to work from. The article explains the methodology employed to locate phrasal items of interest, which partly involved using the Phrases in English (PIE) database. Details are presented of ten specific areas of descriptive meaning where a number of different adjective combinations were found, as well as a set of emphasizing and/or intensifying adjectives with similar characteristics. Possible reasons are suggested for the doubling-up of meaning, and the relevance of the phenomenon to synonym studies is discussed. More than 200 corpus examples are included.

Multiple producible cues do not aid face-name memory: A reverse production effect

Are there effective mechanisms that can be used to remember someone's name? The production effect is a phenomenon that exemplifies memory's robust benefit for studied words or phrases that have been spoken out loud, as opposed to only hearing or seeing them. However, this robust effect has not yet been identified for face-name pairings. The present study seeks to examine the boundary conditions of the production effect in face-name pairings by incorporating the additional cue of valenced adjectives. Participants were presented with facial images and a sentence stating the name and a description of the individual. Sentences were learned in one of four ways: saying the sentence out loud, reading it silently, reading it while hearing it, or only listening to the sentence presented while viewing the face. Memory for the face, name, and adjective combinations were tested using various types of cues: face only (Experiment 1a), or face and name or adjective (Experiment 1b & Experiment 2). Results replicate the lack of a production effect for face-name memory, and instead support a reverse effect for such stimuli. These findings indicate the unique processing of faces and highlight boundary conditions of the production effect.

Using Game Reviews to Recommend Games

We present a recommender system intended to be used by a community of gamers. The system uses free-form text reviews of games written by the members of the community, along with information about the games that a particular user likes, in order to recommend new games that are likely to be of interest to that user. The system uses the frequency of co-occurrence of word pairs that appear in the reviews of a game as features that represent the game. The pairs consist of adjectives and context words; i.e., words that appear close to an adjective in a review. Because of the extremely large number of possible combinations of adjectives and context words, we use information-theoretic co-clustering of the adjective-context word pairs to reduce the dimensionality. Games are represented using the standard information retrieval vector space model, in which vector features are based on the frequency of occurrence of cocluster pairs.We present the results of three experiments with our system. In the first experiment, we use a variety of strategies to relate frequencies of co-cluster pairs to vector features, to see which produces the most accurate recommendations. In the second, we explore the effects of co-cluster dimensionality on the quality of our system’s recommendations. In the third experiment, we compare our approach to a baseline approach using a bag-of-words technique and conclude that our approach produces higher quality recommendations.

Collocational Development during a Stay Abroad

The purpose of the current study was to explore if and how additional-language learners may show changes in phraseological patterns over the course of a stay in a target-language environment. In particular, we focused on noun+adjective combinations produced by a group of additional-language speakers of French at three points in time, spanning 21 months and including an academic year in France. We extracted each combination from a longitudinal corpus and determined frequency counts and two strength-of-association measures (Mutual information [MI] score and Log Dice) for each combination. Separate analyses were conducted for frequency and the strength-of-association measures, revealing that phraseological patterns are significantly predicted by adjective position in the case of all three measures, and that MI scores showed significant change over time. We interpret the results in light of past research that has reported contradictory findings concerning change in phraseological patterns following an immersion experience.

Relative Clause and Downstep in Japanese

It has been proposed that Japanese downstep, in which the pitch register is lowered after an accented phrase, is sensitive to certain syntactic boundaries. In this paper, we investigate whether downstep is blocked at the relative clause boundary in a production experiment with ten speakers. The results suggest that it does not block downstep. On the other hand, there is a difference between adjectives and verbs when they are used attributively with a head noun: Downstep is observed robustly in the verb condition, whereas there is much inter-speaker variation in the adjective condition. Taken together with the results of past research, we propose that the different patterns found by syntactic category, in particular, adjectives, verbs, and nouns, may be explained by assuming speakers’ knowledge of the behavior of these categories that is activated when they pronounce the phrase. Nouns and verbs are readily available as a combined concept in Japanese and thus downstep is not blocked, whereas combinations of adjectives are not so readily available, and thus speakers may insert a boundary, breaking up a phrase that would otherwise constitute a single domain for downstep.

Study on Localizer ‘zhong(中)’ through Analysis of Corpus Data - Focus on Comparison with ‘li(里) ’ -

Among the localizers, this research analyzed corpus data to explain the background, since ‘zhong(中)’ has a characteristic to be combined with verb, adjective, etc.BR While localizer ‘li(裏)’ is combined with noun with spatiality or intensifies spatiality, localizer ‘zhong(中)’ has a lower rate to respond to the verb that strongly requires spatiality, which frequently is combined with the verb that targets the content instead of the container.BR Here, as the result of widely observing ‘N’, which has the structure of ‘N中’, two characteristics to explain the combination of verb and adjective were found. One is, gathering of ‘N’ made of ‘n’, another is, the environment and the state where ‘N’ surrounds the actor. The former can be connected with combination with verb. Considering the action that a verb expresses also can be the gathering of phased motions, it has a semantic continuity with ‘N’ as the concrete object ‘n’, the gathering of ‘n’ which composes the meaning of word, which again becomes a link to share the localizer ‘zhong(中)’.BR Another characteristic of ‘N’ is to be combined with adjective. Relation with the actor and the climatic phenomena such as snow, rain, and wind that surround the actor enable extension of cognition to abstract areas such as the atmosphere that surrounds the actor, hope, and joy, etc., which is connected with the adjective through the link regarding that it is in the state of the actor.BR Based on the above two characteristics, the common point is the ‘density’ of the inner structure of ‘N’. ‘zhong(中)’ has the probability of complex significance compared to the position between ‘inside-outside’ shown in the container, considering the relation with surrounding space that the position ‘center’ premises. Besides the space(空間), based on the existence of container, it implies ‘inside’, ‘internal’, ‘interior’, etc., this research concluded that these are ‘dense assembly’, rather than ‘space’. BR

On Being Negative

This paper investigates the pragmatic expressions of negative evaluation (negativity) in two corpora: (i) comments posted online in response to newspaper opinion articles; and (ii) online reviews of movies, books and consumer products. We propose a taxonomy of linguistic resources that are deployed in the expression of negativity, with two broad groups at the top level of the taxonomy: resources from the lexicogrammar or from discourse semantics. We propose that rhetorical figures can be considered part of the discourse semantic resources used in the expression of negativity. Using our taxonomy as starting point, we carry out a corpus analysis, and focus on three phenomena: adverb + adjective combinations; rhetorical questions; and rhetorical figures. Although the analysis in this paper is corpus-assisted rather than corpus-driven, the final goal of our research is to make it quantitative, in extracting patterns and resources that can be detected automatically.

Extracting features from online software reviews to aid requirements reuse

Sets of common features are essential assets to be reused in fulfilling specific needs in software product line methodology. In Requirements Reuse (RR), the extraction of software features from Software Requirement Specifications (SRS) is viable only to practitioners who have access to these software artefacts. Due to organisational privacy, SRS are always kept confidential and not easily available to the public. As alternatives, researchers opted to use the publicly available software descriptions such as product brochures and online software descriptions to identify potential software features to initiate the RR process. The aim of this paper is to propose a semi-automated approach, known as Feature Extraction for Reuse of Natural Language requirements (FENL), to extract phrases that can represent software features from software reviews in the absence of SRS as a way to initiate the RR process. FENL is composed of four stages, which depend on keyword occurrences from several combinations of nouns, verbs, and/or adjectives. In the experiment conducted, phrases that could reflect software features, which reside within online software reviews were extracted by utilising the techniques from information retrieval (IR) area. As a way to demonstrate the feature groupings phase, a semi-automated approach to group the extracted features were then conducted with the assistance of a modified word overlap algorithm. As for the evaluation, the proposed extraction approach is evaluated through experiments against the truth data set created manually. The performance results obtained from the feature extraction phase indicates that the proposed approach performed comparably with related works in terms of recall, precision, and F-Measure.

Does Temporal Integration Occur for Unrecognizable Words in Visual Crowding?

Visual crowding—the inability to see an object when it is surrounded by flankers in the periphery—does not block semantic activation: unrecognizable words due to visual crowding still generated robust semantic priming in subsequent lexical decision tasks. Based on the previous finding, the current study further explored whether unrecognizable crowded words can be temporally integrated into a phrase. By showing one word at a time, we presented Chinese four-word idioms with either a congruent or incongruent ending word in order to examine whether the three preceding crowded words can be temporally integrated to form a semantic context so as to affect the processing of the ending word. Results from both behavioral (Experiment 1) and Event-Related Potential (Experiment 2 and 3) measures showed congruency effect in only the non-crowded condition, which does not support the existence of unconscious multi-word integration. Aside from four-word idioms, we also found that two-word (modifier + adjective combination) integration—the simplest kind of temporal semantic integration—did not occur in visual crowding (Experiment 4). Our findings suggest that integration of temporally separated words might require conscious awareness, at least under the timing conditions tested in the current study.

On the syntactic status of I 'm sure

This study tests whether the syntactic status of the subject–adjective combination I’m/I am sure is similar to the subject–verb combination I think (i.e., whether it exhibits the same signs of grammaticalisation along two different parameters). More specifically, the study is concerned with the ability of I’m/I am sure to (i) occur in clause-medial and clause-final position, and with (ii) its preference for that-omission, by comparing the behaviour of I’m/I am sure with the results reported for I think in previous studies. The results show that I’m/I am sure behaves in a similar way to I think both in terms of its ability to occur in clause-medial and clause-final position, and in terms of its preference for that-omission. However, sure is both much less frequent than think in general, and is also proportionally less dominant among the class of adjectival predicates followed by that-clauses than think is among verbal predicates. This makes it difficult to argue that they have developed independently through the same frequency correlation. Instead, I argue that sure and think are part of the same grammaticalised constructional schema, and that the frequency of think could be seen to have an impact on the grammatical status of the parallel construction with sure.

Scandinavian pancake constructions as a family of constructions

This paper deals with a classical problem in Scandinavian grammar, so-called ‘pancake sentences’, nicknamed after examples like Pannekaker er godt ‘Pancakes are good’ where there seemingly is disagreement between the plural subject and the predicative adjective in the neuter singular. Our aim is twofold. From the theoretical point of view, we shall argue that there are advantages with a construction-based approach, and that such an approach is superior to previous analyses within various generative frameworks.The main reason for this is that the data require generalizations over combinations of subjects and predicative adjectives at a rather specific level. From a more empirical point of view, we shall argue that Scandinavian displays a range of different, but related pancake constructions. For the first time, corpus data are brought into the debate. We show that a construction type that has not received much attention previously is in fact the most frequent type, namely constructions where the subject is a deverbal noun.

Children’s Knowledge of the Spanish Copulas Ser and Estar with Novel Adjectives

Previous research suggests that children’s behavior with respect to the interpretation of the Spanish copulas ser and estar may differ depending on their knowledge of the world and whether the copulas occur with real adjectives or novel adjectives in the experimental prompt: Whereas 5-year-old children associate estar + real adjective combinations to temporary properties (Schmitt & Miller 2007), children as old as 12 years of age are not adultlike in their interpretation of estar + novel adjective combinations, where novel adjectives are paired with properties for which children already have a Spanish word (Alonqueo 2007). In this article, we present two experimental studies on Puerto Rican children’s use of copulas and show that when novel adjectives are paired with novel properties and those novel adjectives occur in the experimental prompt with the copula estar, children associate the novel adjective to a temporary property by 4 years of age. These results highlight the importance of the copula to the meaning of the copula + adjective combination.

Comprehension of degree modifiers by pre-school children: What does it mean to be ‘a bit cold’?

Abstract Although even young infants were shown to have some understanding of (adjectival) scalarity, studies of children’s spontaneous speech suggest that the acquisition of scalar semantics may not yet be completed by the time children enter primary school. In the present study, this hypothesis is tested by investigating the comprehension of diminishers (‘a bit’) and consequential degree modifiers (‘too’) modifying relative adjectives (long, warm) in a group of 5-year-old Dutch-speaking children. Based on earlier production studies, it is hypothesized that by age 6 children are adult-like in their comprehension of ‘too’ and not yet target-like in the comprehension of ‘a bit’ modifying relative adjectives. The results of the comprehension experiment demonstrate that some children have already acquired the semantics of both ‘too’ and ‘a bit’, whereas others still have trouble understanding combinations of relative adjectives with each of these degree adverbs. Furthermore, poor comprehenders need more time to process sentences with ‘a bit’ compared to the same sentences with ‘too’, presumably revealing a greater conceptual complexity of diminishers. These findings are consistent with the idea that the acquisition of scalarity has a protracted time course

There is no such thing as a free combination: a usage-based study of specific construals in adverb–adjective combinations

The study is aimed at revealing collocational adverb–adjective patterns in the British National Corpus (BNC). The adverbs selected for the study include the maximizers absolutely, completely, entirely, fully, perfectly, totally, utterly, wholly. The study involves searches on both the selected adverbs and the adjectives they modify in a bi-directional fashion. It is claimed that only a cognitive and usage-based approach in terms of underlying conceptual structures can provide an accurate description of collocational patterns. The results show that a large proportion of the adjectives have strong bonds with particular maximizers. This is explained through the basic conceptual structure of Boundedness/Scalarity, i.e. the degree to which the adjective lends itself to a bounded or a scalar construal and the adverb is biased towards a totality construal (which is the kind of construal to be expected from maximizers). The results support the hypothesis that a substantial part of the adverb–adjective combinations investigated are (semi)-prefabricated units, presumably easily accessed by native speakers because the combinations are the result of specific construals and their members have close associative and conceptual links in the mental lexicon.

A Model of Language-Based Impression Formation and Attribution Among Germans

Research on the English and the Japanese languages has shown that impression formation and attribution processes can be modeled as stemming from a desire to maximize affective coherence in linguistic representations of social events. This article describes a replication in the German language, simultaneously uncovering subtle cultural distinctions. Subjects ( N = 1,905) rated 376 nouns, 393 verbs, 331 adjectives, 128 combinations of nouns and adjectives, and 100 short descriptions of social interaction on the three dimensions of the semantic differential (evaluation, potency, and activity). The data were used to estimate a set of regression equations that can be used to model impression formation and attribution. Sample applications of the model demonstrate its ability to predict the outcome of textbook classics in experimental social psychology.