Evaluating Information Credibility of Digital Content using Hybrid Approach

Digital content is not always credible where credibility refers to the believability of the content. Since the Web has both lowered the cost and increased access to information, it has enabled users in acquiring more content from a number of useful sources. For Web users, verifying information without having any prior knowledge is difficult and the same is true for computers because the current structure of Web does not support semantics. The existing solution to measuring Web credibility is either by using computers or humans. However, credibility judgment by users often considered impractical because they are: (i) time consuming, (ii) user require training or (iii) involve cost if experts do the assessment. There is also no unified model when evaluating using computers. This research looks into evaluation done by humans and computers, and identifies factors in order to use them as variables for measuring credibility. The paper highlights the significance of these factors and maps them into categories including accuracy, authority, aesthetics, professionalism, popularity, currency, impartiality and quality. The effectiveness of a hybrid is also discussed that can produce reliable results by integrating important variables from both human and computer judgments. The outcome of this study would significantly change the way users acquire and use information. The effectiveness (relevancy and trustworthiness of information) and efficiency (time taken to obtain the required credible information) will increase and decrease respectively.