Volume 2, Issue 2 (to appear in Digital Library in June, 2012):

Regular Article

• A Computational Framework for Media Bias Mitigation

  By Souneil Park, Seungwoo Kang, Sangyoung Chung, and Junehwa Song

  The bias in the news media is an inherent flaw of the news production process. The bias often causes a sharp increase in political polarization and in the cost of conflict on social issues such as the Iraq war. This paper presents NewsCube, a novel internet news service that aims to mitigate the effect of media bias. NewsCube automatically creates and promptly provides readers with multiple classified views on a news event. As such, it helps readers understand the event from multiple views and formulate their own, more balanced viewpoints. The media bias problem has been studied extensively in mass communication and social science. This paper reviews related mass communication and journalism studies and provides a structured view of the media bias problem and its solution space. We propose media bias mitigation as a practical solution and demonstrate it through NewsCube. We evaluate and discuss the effectiveness of NewsCube through various performance studies.

Special Issue on Personalization and Persuasion

(Special issue editors: Shlomo Berkovsky, Jill Freyne, and Harri Oinas-Kukkonen)

• Influencing Individually: Fusing Personalization and Persuasion

  By Shlomo Berkovsky, Jill Freyne, and Harri Oinas-Kukkonen

  Personalized technologies aim to enhance user experience by taking into account users’ interests and preferences and other relevant information about users. Persuasive technologies aim to modify user’s attitudes, intentions, or behavior through computer-human dialogue and social influence. While both personalized and persuasive technologies influence user interaction and behavior, we posit that this influence can be significantly increased if the two technologies are combined to create personalized and persuasive systems. For example, the persuasive power of a one-size-fits-all persuasive intervention can be enhanced if the system considers the user who is being influenced and his or her susceptibility to the persuasion being offered. Likewise, personalized technologies can cash in on increased success in terms of user satisfaction, revenue, and user experience if their services use persuasive techniques. Hence the coupling of personalization and persuasion has the potential to enhance the impact of both technologies, as is illustrated by the articles in this special issue.

• Adaptive Persuasive Systems: A Study of Tailored Persuasive Text Messages to Reduce Snacking

  By Maurits Kaptein, Boris D. Ruyter, Panos Markopoulos, and Emile Aarts

  This article describes the use of personalized short text messages (SMSs) to reduce snacking. First, we describe the development and validation (N = 215) of a questionnaire to measure individual susceptibility to different social influence strategies. To evaluate the external validity of this Susceptibility to Persuasion Scale (STPS), we set up a 2-week text-messaging intervention which used text messages that implement social influence strategies as a prompt to reduce snacking behavior. The results of this experiment (N = 73) show that messages that are personalized (tailored) to an individual on the basis of his or her scores on the STPS lead to a larger decrease in snacking consumption than randomized messages or messages that are not tailored (contra-tailored) to the individual. We discuss the importance of this finding for the design of persuasive systems and detail how designers can use tailoring at the level of social influence strategies to increase the effectiveness of their persuasive technologies.

• Investigating the Persuasion Potential of Recommender Systems from a Quality Perspective: an Empirical Study

  By Paolo Cremonesi, Franca Garzotto, and Roberto Turrin

  Recommender systems (RSs) help users search large amounts of digital contents and services by allowing them to identify the items that are likely to be more attractive or useful. RSs play an important persuasion role, as they can potentially augment the users’ trust in an application and orient their decisions or actions in specific directions. This paper explores the persuasiveness of RSs, presenting two large-scale empirical studies that address a number of research questions. First, we investigate whether a design property of an RS defined by the statistically measured quality of its algorithms is a reliable predictor of its potential for persuasion. This latter factor is measured in terms of perceived quality, defined by the overall satisfaction, as well as by how users judge the accuracy and novelty of the recommendations. We designed an empirical study involving 210 participants and implemented seven full-sized versions of a commercial RS, each one using the same interface and data set (a subset of the Netflix data set), but each with a different recommender algorithm. In each experimental configuration, we computed the statistical quality (recall and F-measures) and collected data regarding the quality perceived by 30 users. The results show that algorithmic attributes are less crucial than we may expect in determining the user’s perception of an RS’s quality, and they suggest that the user’s judgment and attitude toward a recommender are likely to be more affected by factors related to the user experience. Second, we explore the persuasiveness of RSs in the context of large interactive TV services. We report on a study aimed at determining whether measurable persuasion effects (e.g., changes in shopping behavior) can be achieved through the introduction of a recommender. Our data, collected over more than one year, allow us to conclude that (1) the adoption of an RS can affect both the lift factor and the conversion rate, determining an increased volume of sales and influencing the user’s decision whether to buy one of the recommended products; (2) the introduction of an RS tends to diversify purchases and orient users toward less obvious choices (i.e., the “long tail”); and (3) the perceived novelty of recommendations is likely to be more influential than their perceived accuracy. Overall, the results of these studies improve our understanding of the persuasion phenomena induced by RSs, and they have implications that can be of interest to academic scholars, designers, and adopters of this class of systems.

• System Personality and Persuasion in Human-Computer Dialogue

  By Pierre Y. Andrews

  The human-computer dialogue research field has been studying interaction with computers since the early years of artificial intelligence; but research has often focused on very practical tasks to be completed with the dialogues. A new trend in the field is to implement persuasive techniques with automated interactive agents; unlike booking a train ticket, for example, such dialogues require the system to show more anthropomorphic qualities. The influences of such qualities on the effectiveness of persuasive dialogue are only starting to be studied. In this article, we focus on one important perceived trait of the system – personality – and explore how it influences the persuasiveness of a dialogue system. We introduce a new persuasive dialogue system and combine it with a state-of-the-art personality utterance generator. By doing so, we can control the system’s extraversion personality trait and observe the trait’s influence on the user’s perception of the dialogue and its output. In particular, we observe that a user’s extraversion influences his or her perception of the dialogue and its persuasiveness, and that the perceived personality of the system can affect its perceived trustworthiness and its persuasiveness . We believe that these observations will help to set up guidelines to tailor dialogue systems to the user’s interaction expectations and to improve persuasive interventions.