Improving Judgment Reliability in Social Networks via Jury Theorems
- 275 Downloads
Opinion aggregators—such as ‘like’ or ‘retweet’ counters—are ubiquitous on social media platforms and often treated as implicit quality evaluations of the entry liked or retweeted, with higher counts indicating higher quality. Many such aggregators are poor quality evaluators as they allow disruptions of the conditions for positive wisdom-of-the-crowds effects. This paper proposes a design of theoretically justified aggregators that improve judgment reliability. Interpreting states of diffusion processes on social networks as implicit voting scenarios, we specify procedures for isolating sets of independent voters in order to use jury theorems to quantify the reliability of network states as quality evaluators. As real-world networks tend to grow very large and independence tests are computationally expensive, a primary goal is to limit the number of such tests. We consider five procedures, each trading a degree of reliability for efficiency, the most efficient requiring a low-degree polynomial number of tests.
We thank the reviewers of LORI-VII and the participants of the Social Interactions in Epistemology and in Economics conference (Cph, 29-31/5/2019) for insightful comments. The Center for Information and Bubble Studies is funded by the Carlsberg Foundation. RKR was partially supported by the DFG-ANR joint project Collective Attitude Formation [RO 4548/8-1].
- 10.Condorcet, M.M.d.: Essai sur l’Application de l’Analyse à la Probabilité des Décisions Rendues à la Pluralité des Voix. Paris (1785)Google Scholar
- 18.Dietricht, F., Spiekermann, K.: Jury Theorems. In: Fricker, M., Graham, P.J., Henderson, D., Pedersen, N., Wyatt, J. (eds.) The Routledge Companion to Social Epistemology. Routledge (2019)Google Scholar
- 20.Acharya, J., Daskalakis, C., Kamath, G.: Optimal testing for properties of distributions. In: Cortes, C., Lawrence, N.D., Lee, D.D., Sugiyama, M., Garnett, R. (eds.) Advances in Neural Information Processing Systems 28, pp. 3591–3599. Curran Associates, Inc. (2015)Google Scholar
- 21.Kleinberg, J., Tardos, E.: Algorithm Design. Pearson (2005)Google Scholar