Continuous Top-k Dominating Query of Incomplete Data over Data Streams

Decision support application plays an important role recently. The example of a decision support application is stock market analysis, weather data analysis, sensor network management and others. Recently, the top-k dominating query has become one of the preferred research topics in decision support application. This query returns k number of superior objects which have the highest dominating score in dataset among others. A problem may arise in the data stream environment system that requires to monitor the query result continuously. An efficient method which able to reduce the iteration of the computational process is needed. On the other hand, the real data does not always have a value in each of its attribute or dimension of data. So, unlike the complete data, another solution is required to deal with the query processing task over incomplete data. This paper proposes a solution for obtaining the top-k dominating object in dynamic environment which serves the incomplete data. The event-based method is proposed to handle the continuous top-k dominating query task efficiently. By evaluating the performance over synthetic and real-life data, the proposed solution is proven to have significantly more efficient query computational time compared to the naive one.