A corpus analysis of online news comments using the Appraisal framework

Abstract

We present detailed analyses of the distribution of Appraisal categories (Martin & White, 2005) in a corpus of online news comments. The corpus consists of just over one thousand comments posted in response to a variety of opinion pieces on the website of the Canadian English-language newspaper The Globe and Mail. We annotated all the comments with labels corresponding to different categories of the Appraisal framework. Analyses of the annotations show that comments are overwhelmingly negative and that they favour two of the subtypes of Attitude, Judgement and Appreciation. The paper contributes a methodology for annotating Appraisal, examining the interaction of Appraisal with negation, the constructive nature of comments, and the level of toxicity found in them. The results show that highly opinionated language is expressed as opinion (Judgement and Appreciation) rather than as an emotional reaction (Affect). This finding, together with the interplay of evaluative language with constructiveness and toxicity in the comments, can be applied to the automatic moderation of online comments.

Attitudes Towards Global Warming on Twitter

“…Twitter users tended to express their opinions on global warming through Appreciation (evaluating things), and Judgement (evaluating people's behaviours), rather than directly expressing their emotions through Affect. As Cavasso and Taboada (2021) put it, "strong emotional content is couched in terms of Judgement and Appreciation" (p. 27). Put another way, the results suggested that language users produce sentences like "Global warming is real/disaster" or "It's crazy that ignorant people just don't believe in global warming" rather than "I'm angry/worried about global warming".…”

Discourse Analysis of the 2022 Australian Tennis Open: A Multimodal Appraisal Perspective

“…The term cluster refers to the local groupings of items, be they words, sounds, images or video frames. Then emphasis was given to minimality and contextuality, as described by Cavasso and Taboada (2021). 'The principle of minimality means that the item to be annotated (cluster-span) should be as short as possible, while at the same time including all the words that convey Attitude.…”

“…Appraisal Theory deals with how evaluative language aligns readers according to common values and attitudes. Since 2005, Appraisal has been used to: analyse static multimodal texts (Economou 2009, Swain 2012, Mills & Stone et al 2020; it has also been applied to online discourse (language) and combined with corpus techniques (Cavasso & Taboada 2021, Drasovean & Tagg 2015; recently a model has also been proposed for paralanguage that incorporates Appraisal (Martin & Zappavigna 2019;Ngo, Hood et al 2021). To the best knowledge of this researcher, no studies have dealt with a multimodal discourse Appraisal analysis of a corpus of online texts.…”

Characterising Online News Comments: A Multi-Dimensional Cruise Through Online Registers

“…This description is also in line with our other recent analyses. Ehret and Taboada (2020) compared online news comments to traditional written and spoken registers and found that they are strongly evaluative in nature, combining argumentative, informational, and some involved features (Ehret and Taboada, 2020, 23), while Cavasso and Taboada (2021) observe their overwhelmingly negative nature, with personal affective opinion (I hate the candidate) eschewed in favour of more detached evaluation (The candidate is incompetent; The candidate's policies are bad). As illustrated in (4), online news comments can thus range from involvedevaluative to involved-argumentative and informationalargumentative.…”