Detecting Expressions of Blame or Praise in Text

Detecting Expressions of Blame or Praise in Text Udochukwu Orizu and Yulan He School of Engineering and Applied Science Aston University, UK orizuus@aston.ac.uk, y.he@cantab.net Abstract The growth of social networking platforms has drawn a lot of attentions to the need for social computing. Social computing utilises human insights for computational tasks as well as design of systems that support social behaviours and interactions. One of the key aspects of social computing is the ability to attribute responsibility such as blame or praise to social events. This ability helps an intelligent entity account and understand other intelligent entities’ social behaviours, and enriches both the social functionalities and cognitive aspects of intelligent agents. In this paper, we present an approach with a model for blame and praise detection in text. We build our model based on various theories of blame and include in our model features used by humans determining judgment such as moral agent causality, foreknowledge, intentionality and coercion. An annotated corpus has been created for the task of blame and praise detection from text. The experimental results show that while our model gives similar results compared to supervised classifiers on classifying text as blame, praise or others, it outperforms supervised classifiers on more finer-grained classification of determining the direction of blame and praise, i.e., self-blame, blame-others, self-praise or praise-others, despite not using labelled training data. Keywords: blame/praise detection, text classification, social computing. 1. Introduction With the prevalence of social networking sites, the need for social computing is constantly growing. Social computing utilises human insights for computational tasks as well as design of systems that support social behaviours and inter- actions (Parameswaran et al, 2007). In recent years, so- cial computing is becoming key in many research areas and technological systems such as learning, human-computer interaction, entertainment and many more. One of the key aspects of social computing is the ability of an entity to infer the social behaviour of not just itself but of other en- tities as well (Parameswaran et al, 2007; Mao et al, 2011). This inference includes the ability to pass judgement and determine if an entity is blameworthy or praiseworthy and allocate blame or praise where appropriate (Tognazzini et al, 2014). Blame and praise are closely related, to blame an entity is to hold that entity morally responsible for doing something of a negative outcome while praise is to hold that entity morally responsible for doing something of a posi- tive outcome (Eshleman, 2014) . Detection of blame/praise can be used in a variety of applications such as identifying entities holding moral responsibilities in multi-agent sys- tems, helping with the detection of emotions such as guilt, remorse, admiration, and shame, and many more (Ortony, 1990). In this paper, we focus on detecting blameworthiness and praiseworthiness based on the “Path Model of Blame” pre- sented in (Malle et al, 2014). In particular, we propose an approach by adapting the original Path Model of Blame and combine natural language processing techniques for the de- tection of blame or …

Content truncated to 500 words