Structuring the Processing Frameworks for Data Stream Evaluation and Application

TL;DR

This paper proposes a structured framework for evaluating data stream processing methods, emphasizing realistic constraints like delayed labels and concept drift, and introduces a taxonomy linking drift detection and classification techniques.

Contribution

It introduces a taxonomy of data stream processing frameworks that accounts for label delay and concept drift, improving evaluation reliability in realistic scenarios.

Findings

Reviewed current data stream processing methods.

Verified outcomes in simulated environments.

Proposed a taxonomy linking drift detection and classification.

Abstract

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.