This paper investigates a little-studied class of adjectives that we refer to as ‘complex adjectives’, i.e., operationally, adjectives constituted of at least two word tokens separated by a hyphen. We study the properties of these adjectives using two very large text collections: a portion of Wikipedia and a Web corpus. We consider three corpus-based measures of morphological productivity, and ...

Gradable adjectives are inherently vague and are used by clinicians to document medical interpretations (e.g., severe reaction, mild symptoms). We present a comprehensive study of gradable adjectives used in the clinical domain. We automatically identify gradable adjectives and demonstrate that they have a substantial presence in clinical text. Further, we show that there is a specific pattern ...

The literature on predicates of personal taste (PPTs) has focused on a surprisingly limited range of expressions. This narrow focus has led to problematic theoretical conclusions about the syntax and semantics. I argue that PPTs are an instance of a more interesting general category of (what I call) evaluational predicates. Like other gradable adjectives, PPTs are sensitive to a relevant degree...

We compare several different corpusbased and lexicon-based methods for the scalar ordering of adjectives. Among them, we examine for the first time a lowresource approach based on distinctivecollexeme analysis that just requires a small predefined set of adverbial modifiers. While previous work on adjective intensity mostly assumes one single scale for all adjectives, we group adjectives into d...

We examine predicative adjectives as an unsupervised criterion to extract subjective adjectives. We do not only compare this criterion with a weakly supervised extraction method but also with gradable adjectives, i.e. another highly subjective subset of adjectives that can be extracted in an unsupervised fashion. In order to prove the robustness of this extraction method, we will evaluate the e...

Comparison of positive and negative naturally read adjectives to neutral adjectives yielded an overlapping higher BOLD response in the occipital and the orbitofrontal cortex (gyrus rectus). Superior medial frontal gyrus and posterior cingulate gyrus showed higher BOLD response to negative adjectives and inferior frontal gyrus to positive adjectives. The overlap of activated regions and lack of ...

The goal of this research is to automatically predict human judgments of document qualities such as subjectivity, verbosity and depth. In this paper, we explore the behavior of adjectives as indicators of subjectivity in documents. Specifically, we test whether a subset of automatically derived subjective adjectives (Wiebe, 2000b), selected a priori, behaves differently than other adjectives. 3...

Female students rated males characterized by a photograph and 2 personalitytrait adjectives in terms of desirability as a date. The experimental design tested how these 3 component pieces of information were integrated to form the overall impression. The results supported the hypothesis that the 2 adjectives were simply averaged together, but there was a strong interaction between adjectives an...

This paper examines the significance of the schematic domain of BOUNDEDNESS in adjectives. It is proposed that boundedness in adjectives is a fundamental characteristic associated with gradability. Cross-categorial correspondences are made to nouns and verbs, where boundedness is a feature of countability and aktionsart respectively. Two basic types of gradable adjectives are distinguished: tho...

Web users produce more and more documents expressing opinions. Because these have become important resources for customers and manufacturers, many have focused on them. Opinions are often expressed through adjectives with positive or negative semantic values. In extracting information from users’ opinion in online reviews, exact recognition of the semantic polarity of adjectives is one of the m...