Fuzzy Classification of the Flow of Events for Decision-Making in Smart Systems

Decision making engine based on the classification of the data stream in real time is the main challenge of the Smart City Application (SCA). Fuzzy logic system (FLS) can be used as such engine if the data stream dimension will be significantly reduced. The two-stage Computing with Words (CWW) approach overcomes the dimensionality problem by using abstraction engine at the first stage, which maps the meaning of stream data from sensors to the meaning of a few words. In this article the CWW approach is extended by a Short-Term Memory (STM) model that stores a time sequence of events in the form of fuzzy characteristics of words of a high level of abstraction. The STM footprint model is supported by the footprint blur algorithm, in which calculations are performed with blurring of information about events with time expiration. The classification is proposed to be performed by a fuzzy assessment of the proximity of the STM footprint and the prototype of the class of the flow of events. An example of the use of fuzzy classification of the flow of events model is given.