The paper conceptualizes the societal impacts of disinformation in hopes of developing a computational approach that can identify disinformation in order to strengthen social resilience. An innovative approach that considers the sociotechnical interaction phenomena of social media is utilized to address and combat disinformation campaigns. Based on theoretical inquiries, this study proposes conducting experiments that capture subjective and objective measures and datasets while adopting machine learning to model how disinformation can be identified computationally. The study particularly will focus on understanding communicative social actions as human intelligence when developing machine intelligence to learn about disinformation that is deliberately misleading, as well as the ways people judge the credibility and truthfulness of information. Previous experiments support the viability of a sociotechnical approach, i.e., connecting subtle language-action cues and linguistic features from human communication with hidden intentions, thus leading to deception detection in online communication. The study intends to derive a baseline dataset and a predictive model and by that to create an information system artefact with the capability to differentiate disinformation.