A data centric approach to utility change detection in online social media

This paper considers the problem of detecting changes in utility maximizing behaviour of agents in online social media. Such changes in utility maximizing behaviour in online social media occur due to the effect of marketing, advertising, or changes in ground truth. In contrast to traditional signal processing techniques, our approach is data-centric. We use the framework of revealed preference to detect the unknown time point (change point) at which the utility function changed. We derive necessary and sufficient conditions for detecting the change point. In addition, we provide an algorithm to recover the utility function before and after the change point. The results developed are illustrated on the Yahoo! Tech Buzz dataset. From the dataset, we obtain the following useful insights: First, the changes in ground truth affecting the utility of the agent can be detected by utility maximization behaviour in online search. Second, the recovered utility functions satisfy the single crossing property indicating strategic substitute behaviour in online search.