Frameworks for Collective Intelligence

Abstract

Over the last few years, Collective Intelligence (CI) platforms have become a vital resource for learning, problem solving, decision-making, and predictions. This rising interest in the topic has to led to the development of several models and frameworks available in published literature. Unfortunately, most of these models are built around domain-specific requirements, i.e., they are often based on the intuitions of their domain experts and developers. This has created a gap in our knowledge in the theoretical foundations of CI systems and models, in general. In this article, we attempt to fill this gap by conducting a systematic review of CI models and frameworks, identified from a collection of 9,418 scholarly articles published since 2000. Eventually, we contribute by aggregating the available knowledge from 12 CI models into one novel framework and present a generic model that describes CI systems irrespective of their domains. We add to the previously available CI models by providing a more granular view of how different components of CI systems interact. We evaluate the proposed model by examining it with respect to six popular, ongoing CI initiatives available on the Web.

A Survey of Collaborative Reinforcement Learning: Interactive Methods and Design Patterns

“…Arzate and Chirstian presented a survey on the design principles and open challenges of interactive reinforcement learning [6]. Suran and Shweta presented a review on collective intelligence [60]. It can be seen that this research direction is receiving more and more attention from the community.…”

Social Computing and Privacy

“…This approach is evolving significantly with the dissemination of new technologies for accessing, processing, and sharing knowledge, providing resources for training, decision making and forecasting for the development of processes and situations. This requires the development of appropriate models for solving emerging problems, and in this respect, a systematic review of the frameworks for the development of CI (proposed in the scientific literature after year 2000) is made in [7]. As a result of this study, a summary of available knowledge has been made, and based on several popular CI initiatives in the global network, a new model for describing CI systems with clearer presentation of the various components has been proposed.…”

“…The literature increasingly argues that human online behavior often deviates from standard economic patterns of behavior, and this may indicate an impact on optimal system design. As mentioned above [7,10], a dialogue between theoretical and empirical research is needed to define appropriate models. sharing, blogs and micro-blogs, virtual sites and etc.…”

Clustering spatio-temporal bi-partite graphs for finding crowdsourcing communities in IoMT networks

“…Crowdsourcing has been a phenomenon characterised by many overlapping idiosyncratic definitions coming from research fields in volunteered geographic information Senaratne, Mobasheri, Ali, Capineri, & Haklay, 2017), crowd wisdom (Davis-Stober, Budescu, Dana, & Broomell, 2014), collective intelligence (Suran, Pattanaik, & Draheim, 2020), mobile crowd sensing (Guo, Yu, Zhou, & Zhang, 2014;Kazemi et al, 2013;Ogie, 2016), and participatory sensing (Burke, Estrin, Hansen, Parker, Ramanathan, Reddy, Srivastava, 2006;Farkas, 2020); and to date, there is still an absence of shared scientific terms and concepts. The applications of crowdsourcing have been numerous and varied, cutting across a wide range of areas for supporting public processes in planning, decision-making and policy formation activities in public health, education, public safety, environmental monitoring, transportation, and disaster management (Feese, Burscher, Jonas, & Tröster, 2014;Hossain & Kauranen, 2015;Klonner, Marx, Usón, Porto de Albuquerque, & Höfle, 2016;Lashkari, Rezazadeh, Farahbakhsh, & Sandrasegaran, 2019;.…”