Parallel Skyline Query Processing of Massive Incomplete Activity-Trajectories Data

The big spatial temporal data captured from technology tools produce massive amount of trajectories data collected from GPS devices. The top-k query was proposed by many researchers, on which they used distance and text parameters for processing. However, the information related to text parameter like activity is always not presented due to some reason like lack internet connection. Furthermore, with massive amount of keyword semantic activity-trajectories, user may enter the wrong activity to find its activity-trajectory. Therefore, it’s hard to return the desirable results based on the exact keyword activity. Our previous work proposed an efficient algorithm to handle the trajectory fuzzy problem based on edit distance and activity weight. However, the algorithm proposed does not work with incomplete Trajectory DataBases (TDBs). Therefore, the present investigation focuses on handling the trajectory skyline problem based on distance and frequent activities in incomplete TDB. To accelerate the query processing, the massive trajectory objects is managed through Distributed Mining Trajectory R-Tree (DMTR-Tree index) based on R-tree indexes and inverted lists. Afterward, an efficient algorithm is developed to handle the query. For a rapid computation, a cluster-computing framework of Apache Spark with MapReduce is used. Theoretical analysis and the experimental results show a well agreement and both attest on the higher efficiency of the proposed algorithm.