Incorporating expert judgment for detecting eelevant factors in social networks undetected by ordinary methods

Information and communications technology (ICT) has potential to complement information sharing bureaus (ISB) Most companies use social networks as communication channels because they can provide significant business benefits. This paper focuses on the impact of social networks in a Spanish foundation for innovation and knowledge dissemination, and how they affect its main events and activities. We examine the factors underlying a re-tweet on Twitter or a share on Facebook in order to analyze reporting of this foundation's principal events. Comparisons with three statistical models were performed (standard regression and Bayesian regression with non-informative and informative priors). We conclude that the advantage offered by Bayesian over classic methodology is demonstrated by incorporation of collateral information, usually provided by experts, which can refine the model and obtain conclusions that cannot be identified otherwise. This conclusion may have significant implications for companies that make use of social networks.