Efficient Top-K Continuous Query Processing Over Sliding Window Model (SWM) Method on Uncertain Data Stream

Abstract

Query processing using the Uncertain Data Stream (UDS) can be complex in many technological scenarios due to inconsistencies, unclear information, and interpretation latency. As a result of both the sheer amount of data generated and the rate of change, traditional processing methods are in dire need of an upgrade. UDS consists of a finite set of states known as possible worlds (PW), and enhancing data organization can lead to more accurate extraction of user preferences. The number of possible world instances in UDS grows exponentially, making achieving Top-k query processing quickly a significant challenge. Different methods are available to handle Top-k queries in various types of UDS, and their key concerns include reducing duplicate scans of the entire dataset, enhancing uncertainty computation, and focusing on processing the latest tuple item entry. It appears that there have been limited studies conducted on the issue of UDS using the Sliding Window Model (SWM). The current approach for handling continuous queries on UDS within the SWM has proven to be ineffective, resulting in complex trade-offs between maximizing probability and generating high-scoring result sets. The challenge is to find the correct result list that satisfies a Top-k query predicate with scoring and probability. This study proposes a framework for processing Top-k queries for UDS using the sliding window model to improve efficiency. The study also discusses an improved optimization method for reducing computational redundancy in the context of the sliding window model and Top-k query processing. Overall, this research will significantly contribute to the Top-k computational query processing field.