Adaptive Information Sharing with Ontological Relevance Computation for Decentralized Self-Organization Systems

Abstract

Decentralization is a peculiar characteristic of self-organizing systems such as swarm intelligence systems, which function as complex collective responsive systems without central control and operates based on contextual local coordination among relatively simple individual systems. The decentralized particularity of self-organizing systems lies in their capacity to spontaneously respond to accommodate environmental changes in a cooperative manner without external control. However, if members cannot obtain observations of the state of the whole team and environment, they have to share their knowledge and policies with each other through communication in order to adapt to the environment appropriately. In this paper, we propose an information sharing mechanism as an independent decision phase to improve individual members’ joint adaption to the world to fulfill an optimal self-organization in general. We design the information sharing decision analogous to human information sharing mechanisms. In this case, information can be shared among individual members by evaluating the semantic relationship of information based on ontology graph and their local knowledge. That is, if individual member collects more relevant information, the information will be used to update its local knowledge and improve sharing relevant information by measuring the ontological relevance. This will enable more related information to be acquired so that their models will be reinforced for more precise information sharing. Our simulations and experimental results show that this design can share information efficiently to achieve optimal adaptive self-organizing systems.

Exploring the Role of HRM in Facilitating Collaboration and Knowledge Sharing in Decentralized Organizational Structures

“…Decentralized systems benefit from technology that supports adaptive information sharing, enhancing self-organization within teams (Liu et al, 2021). Additionally, the emergence of informal interpersonal networks facilitated by technology plays a vital role in providing access to expertise in highly decentralized systems (Binz-Scharf et al, 2011).…”

An Adaptive Early Fault Detection Model of Induced Draft Fans Based on Multivariate State Estimation Technique

“…If the current operating conditions have a large change and deviate from the operating condition described by the memory matrix, the MSET model will exhibit a decline prediction accuracy, which is called model degradation. Model degradation is a crucial problem of the practical use of most data-driven models [30][31][32]. To maintain a high prediction accuracy, it is necessary to absorb operating information from newly acquired data and achieve the model update [33,34].…”