Online reputation is an invaluable asset for modern organizations as it can help in business performance especially in sales and profit. However, if we are not aware of our reputation, it is difficult to maintain it. Thus, social media analytics is a new tool that can provide online reputation monitoring in various ways such as sentiment analysis. As a result, numerous large-scale organizations have implemented Online Reputation Monitoring (ORM) systems. However, this solution is not supposed to be exclusively for high-income organizations, as many organizations regardless sizes and types are now online. This research attempts to propose an affordable and reliable ORM system using combination of open source analytics tools for both novice practitioners and academicians. We also evaluate its prediction accuracy and we discovered that the system provides acceptable predictions (sixty percent accuracy) and demonstrate a tally prediction of major polarity by human annotation. The proposed system can help in supporting business decisions with flexible monitoring strategies especially for organization that want to initiate and administrate ORM themselves at low cost.

During the current mobile technology era, social media rumors (SMRs) propagate significantly faster, hence, are difficult to control. According to psychological research, SMR propagation can be better controlled with refutation. However, refutation must be timely, and supported with credible information. This is in line with the guideline provided in the Holy Quran, especially verse four of Surat An-Nisa, which urges people to refer to the leader or the authority to ascertain the veracity of new information concerning public security. This paper aims to review prominent prior psychology literature on rumor quelling strategies (RQS), and then maps them with solutions proposed by computer science research. We discovered that computer science studies in the related literature proposed noble solutions aligned with the ultimate goals in psychological research, including refutation. However, we suggest that these solutions be merged, and human intervention cannot be ignored, especially the verification process. Consequently, we identified six processes to be consolidated in a framework to further improve the refutation of SMRs. The six processes are: (1) filteration, (2) topic classification, (3) verifier identification, (4) notification to verifier, (5) verification, and (6) refutation. The identified steps act as the basis for subsequent research progress. � 2017 American Scientific Publishers All rights reserved.