Automatic Expansion of Domain-Specific Affective Models for Web Intelligence Applications

Abstract

Sentic computing relies on well-defined affective models of different complexity—polarity to distinguish positive and negative sentiment, for example, or more nuanced models to capture expressions of human emotions. When used to measure communication success, even the most granular affective model combined with sophisticated machine learning approaches may not fully capture an organisation’s strategic positioning goals. Such goals often deviate from the assumptions of standardised affective models. While certain emotions such as Joy and Trust typically represent desirable brand associations, specific communication goals formulated by marketing professionals often go beyond such standard dimensions. For instance, the brand manager of a television show may consider fear or sadness to be desired emotions for its audience. This article introduces expansion techniques for affective models, combining common and commonsense knowledge available in knowledge graphs with language models and affective reasoning, improving coverage and consistency as well as supporting domain-specific interpretations of emotions. An extensive evaluation compares the performance of different expansion techniques: (i) a quantitative evaluation based on the revisited Hourglass of Emotions model to assess performance on complex models that cover multiple affective categories, using manually compiled gold standard data, and (ii) a qualitative evaluation of a domain-specific affective model for television programme brands. The results of these evaluations demonstrate that the introduced techniques support a variety of embeddings and pre-trained models. The paper concludes with a discussion on applying this approach to other scenarios where affective model resources are scarce.

Integrating Economic Theory, Domain Knowledge, and Social Knowledge into Hybrid Sentiment Models for Predicting Crude Oil Markets

“…The survey identifies six large communities (social media, ML, NLP, opinion mining, Arabic, semi-supervised learning) and their key research topics. Domain-specific affective models [20] go beyond classic models by incorporating application-specific affective categories and interpreting them based on the situational context. An affective model for benchmarking TV shows, for example, might consider fear and sadness to be desirable associations rather than undesirable ones.…”

Adapting Data-Driven Research to the Fields of Social Sciences and the Humanities

“…The selection of the knowledge extraction techniques was guided by the research framework discussed in Section 4, which requires (i) the analysis and tracking of major stakeholders, (ii) the identification of dominant issues, and (iii) the classification of media criticism as either optimistic-constructive or negative. Weichselbraun et al [37] discussed the use of domain-specific affective models, which support capturing emotions that go beyond standard sentiment and emotion models. An assessment that compared the optimisticconstructive and negative dimensions from communication science literature with standard sentiment polarity (dimensions: positive and negative) concluded that, for the purpose of the joint research project, the use of sentiment polarity is an efficient (availability of high-quality sentiment lexicons) and effective (sufficiently high correlation between both metrics) strategy.…”

The Digital Transformation: The Forty Years of Marketing Communication Research

“…The significance of marketing communication has surged, reflecting the market's increasing complexity and consumers' evolving expectations. This field is central to establishing, nurturing, and fortifying the relationship between companies and their customers through a blend of diverse tools, strategies, and tactics, allowing companies to highlight their strengths, narrate their stories, and position their offerings in a competitive marketplace (Weichselbraun, Steixner, Braşoveanu, Scharl, Göbel & Nixon, 2022). Historically, marketing communication strategies like traditional advertising, including billboards, print ads, and radio spots, were predominant.…”