An Effective Pruning Scheme for Top-k Dominating Query Processing on Uncertain Data Streams
Chuan-Chi Lai, Chih-Cheng Fan, Chuan-Ming Liu · 2022
In the modern age of information explosion, everyone can easily obtain all kinds of data, so how to find the most valuable information in massive data has become an important issue. In general, most data collected from the applications of Internet of Things (IoT) become uncertain since there is a probability or part of the data is missing. However, the calculation of the uncertain data will be much more complicated than certain (or deterministic) data. As a result, the performance of uncertain data handling becomes a significant challenge in meeting low latency requirements. In this work, we propose a distributed computing algorithm and apply it to an edge computing to calculate the probabilistic top-k dominating (PTKD) objects of uncertain data. The overall latency of PTKD query processing is significantly reduced. The main idea of this method is to reduce the cost of time without unnecessary calculations of objects. Experiments show that the proposed algorithm can improve 58% latency on average. With a high pruning rate, performance can be reduced by up to 92%.