Structuring the Processing Frameworks for Data Stream Evaluation and Application
This work addresses the need for more reliable evaluation frameworks for data stream processing in real-world applications, but it appears incremental as it builds on existing methods and simulations.
The paper tackled the problem of evaluating data stream classification methods under realistic constraints like delayed and limited label access, resulting in a proposed taxonomy of data stream processing frameworks that links drift detection and classification methods.
The following work addresses the problem of frameworks for data stream processing that can be used to evaluate the solutions in an environment that resembles real-world applications. The definition of structured frameworks stems from a need to reliably evaluate the data stream classification methods, considering the constraints of delayed and limited label access. The current experimental evaluation often boundlessly exploits the assumption of their complete and immediate access to monitor the recognition quality and to adapt the methods to the changing concepts. The problem is leveraged by reviewing currently described methods and techniques for data stream processing and verifying their outcomes in simulated environment. The effect of the work is a proposed taxonomy of data stream processing frameworks, showing the linkage between drift detection and classification methods considering a natural phenomenon of label delay.