Affiliations: [a] Division of Knowledge and System Engineering for ICT, Ton Duc Thang University, Ho Chi Minh City, Vietnam | [b] Faculty of Information Technology, Ton Duc Thang University, Ho Chi Minh City, Vietnam | [c] Faculty of Computer Science, University of Information Technology, Vietnam National University Ho Chi Minh City (VNU-HCM), Vietnam | [d] Department Sistemas Informáticos y Computación, Universidad Complutense de Madrid, Spain
| [e] Faculty of Computer Science and Management, Wroclaw University of Science and Technology, Poland
| [f] Faculty of Information Technology, Nguyen Tat Thanh University, Ho Chi Minh city, Vietnam
Abstract: Recently, the use of the wisdom of crowds (WoC) for finding solutions to a wide range of real-life problems has dramatically expanded. Prior studies have revealed that diversity, independence, decentralization, and aggregation are the determinants of collective wisdom. However, these findings are often based on the so-called point estimates - single values are used as the representations of individual predictions on the task of estimating unknown quantities or predicting outcomes of future events. In some situations, interval values, which are often called interval estimates, can be used for such representations. Accordingly, one can provide an individual prediction in the form of an interval value including a lower and an upper bounds. Taking into account this kind of representation, in this paper, we present a case study in which collectives of randomly selected predictions can outperform those of most accurate predictions. Then, we evaluate the WoC level by taking into account diversity and cardinality. The computational experiments have indicated that diversity is positively related to collective wisdom. Finally, we discuss some related theoretical and practical implications for further research.
Keywords: Collective intelligence, wisdom of crowds, interval estimates
