Abstract:
Credibility is a perceived quality and is evaluated with at least two major components: trustworthiness and expertise. Weblogs (or blogs) are a potentially fruitful genre for exploration of credibility assessment due to public disclosure of information that might reveal trustworthiness and expertise by webloggers (or bloggers) and availability of audience evaluations. The objectives of the planned exploratory study are to compile a list of factors that users take into account in credibility assessment of blog sites, order them in terms of users’ perceived importance, and determine which factors can be recognized and evaluated with Natural Language Processing (NLP) techniques. With partial automation in mind, we propose an analytical framework for blog credibility assessment based on four profile factors: 1) the blogger’s expertise and the amount of offline identity disclosure, 2) the blogger’s trustworthiness (or the overtly stated value system including beliefs, goals, and values), 3) information quality, and 4) appeals of a personal nature. We describe a multi-stage study that combines a qualitative study of credibility judgments of blog-readers with NLP-based analysis of blogs. The study will elicit and test credibility assessment factors (Phase I), perform NLP-based blog profiling (Phase II), and content-analyze blog-readers’ comments for partial profile matching (Phase III).