1: procedure Lookup(σ, k) 2: Z = Get-Top-SS(SSn, k);
3: for each streak z inZ do
4: Q= [|z| · σ, +∞] × [0, z.i) × [z.v · σ, +∞];
5: HM(z) = BBS(R-tree, Q), and relevant data sequence information;
Fig. 9. SOIA-MS versus BIA-MS.
the five real datasets, only D5, the ground motion data sequence from seismic stations, has multiple sequences (from multiple stations). So we construct five real multisequence datasets as follows:
(1) MS1: KMNB, YHNB;
(2) MS2: KMNB, YHNB, YULB, TWGB;
(3) MS3: KMNB, YHNB, YULB, TWGB, TPUB, SSLB;
(4) MS4: KMNB, YHNB, YULB, TWGB, TPUB, SSLB, NACB, YOJ; and (5) MS5: KMNB, YHNB, YULB, TWGB, TPUB, SSLB, NACB, YOJ, HK, HK0, where each code (e.g., KMNB) above denotes one station name.
A.3.1 Overall Comparison. We first look at the overall performance of SOIA-MS and BIA-MS when the full datasets are available. Figure9shows the comparison results that we evaluate them in terms of their (1) index building time (maintenance time), (2) query time (lookup time), and (3) index space.
A.3.2 Historic Moment Exploration with Data Update. We evaluate the performance of SOIA-MS and BIA-SOIA-MS for maintaining the index structure online. In this experiment, we regard the first 98% of a dataset as the initial dataset; i.e., each data sequence in the dataset has 98% as initial, and the index structure of it has been built by a maintenance procedure already.
Next, we examine the performance of SOIA-MS and BIA-MS regarding a data append of the last x% of each data sequence in the dataset, where x= 0.1, 0.5, 1, and 2. Figure10shows the experiment results. While SOIA-MS spends more effort in maintaining minimality, SOIA-MS is much more efficient when looking up the historic moments, as SOIA-MS always maintains a much smaller index size.
The basic principle of SOIA-MS is similar to SOIA, and the impact of parameters σ and k for SOIA-MS is also similar to SOIA. So here we don’t repeat the experiments of parameter sensitivity.
Fig. 10. SOIA-MS versus BIA-MS under data update.
REFERENCES
[1] Foto N. Afrati, Paraschos Koutris, Dan Suciu, and Jeffrey D. Ullman. 2015. Parallel skyline queries. Theory of Comput-ing Systems 57, 4 (2015), 1008–1037.
[2] Gail M. Atkinson and David M. Boore. 2006. Earthquake ground-motion prediction equations for eastern North Amer-ica. Bulletin of the Seismological Society of America 96, 6 (2006), 2181–2205.
[3] Wolf-Tilo Balke, Ulrich Güntzer, and Jason Xin Zheng. 2004. Efficient distributed skylining for web information systems. In International Conference on Extending Database Technology. Springer, 256–273.
[4] Stephan Borzsony, Donald Kossmann, and Konrad Stocker. 2001. The skyline operator. In Proceedings of the 17th International Conference on Data Engineering. IEEE, 421–430.
[5] Jan Chomicki, Parke Godfrey, Jarek Gryz, and Dongming Liang. 2005. Skyline with presorting: Theory and optimiza-tions. In Intelligent Information Processing and Web Mining. Springer, 595–604.
[6] Sarah Cohen, James T. Hamilton, and Fred Turner. 2011. Computational journalism. Communications of the ACM 54, 10 (October 2011), 66–71. DOI:https://doi.org/10.1145/2001269.2001288
[7] Naeemul Hassan, Afroza Sultana, You Wu, Gensheng Zhang, Chengkai Li, Jun Yang, and Cong Yu. 2014. Data in, fact out: Automated monitoring of facts by FactWatcher. In Proceedings of the 40th International Conference on Very Large Data Bases. VLDB Endowment.
[8] Xiao Jiang, Chengkai Li, Ping Luo, Min Wang, and Yong Yu. 2011. Prominent streak discovery in sequence data.
In Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 1280–1288.
[9] Christos Kalyvas and Theodoros Tzouramanis. 2017. A survey of skyline query processing. arXiv preprint arXiv:1704.01788 (2017).
[10] Donald Kossmann, Frank Ramsak, and Steffen Rost. 2002. Shooting stars in the sky: An online algorithm for skyline queries. In Proceedings of the 28th International Conference on Very Large Data Bases. VLDB Endowment, 275–286.
[11] Ming-Yen Lin, Chao-Wen Yang, and Sue-Chen Hsueh. 2016. Efficient computation of group skyline queries on MapRe-duce. GSTF Journal on Computing (JoC) 5, 1 (2016), 69–76.
[12] Dimitris Papadias, Yufei Tao, Greg Fu, and Bernhard Seeger. 2003. An optimal and progressive algorithm for skyline queries. In Proceedings of the 2003 ACM SIGMOD International Conference on Management of Data. ACM, 467–478.
[13] Dimitris Papadias, Yufei Tao, Greg Fu, and Bernhard Seeger. 2005. Progressive skyline computation in database sys-tems. ACM Transactions on Database Systems 30, 1 (March 2005), 41–82. DOI:https://doi.org/10.1145/1061318.1061320 [14] Afroza Sultana, Naeemul Hassan, Chengkai Li, Jun Yang, and Cong Yu. 2014. Incremental discovery of prominent
situational facts. In Proceedings of the IEEE 30th International Conference on Data Engineering. IEEE.
[15] Kian-Lee Tan, Pin-Kwang Eng, and Beng Chin Ooi. 2001. Efficient Progressive Skyline Computation. In Proceedings of the 27th International Conference on Very Large Data Bases (VLDB’01), Peter M. G. Apers, Paolo Atzeni, Stefano Ceri, Stefano Paraboschi, Kotagiri Ramamohanarao, and Richard Thomas Snodgrass (Eds.). Morgan Kaufmann Publishers Inc., San Francisco, CA, 301–310.
[16] Wikipedia. 2017. 2016 Taiwan earthquake — Wikipedia, The Free Encyclopedia. Retrieved fromhttps://en.wikipedia.
org/w/index.php?title=2016_Taiwan_earthquake&oldid=766638710(accessed February 27, 2017).
[17] You Wu, Pankaj K. Agarwal, Chengkai Li, Jun Yang, and Cong Yu. 2012. On one of the few objects. In Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 1487–1495.
[18] Yidong Yuan, Xuemin Lin, Qing Liu, Wei Wang, Jeffrey Xu Yu, and Qing Zhang. 2005. Efficient computation of the skyline cube. In Proceedings of the 31st International Conference on Very Large Data Bases. VLDB Endowment.
[19] Gensheng Zhang, Xiao Jiang, Ping Luo, Min Wang, and Chengkai Li. 2014. Discovering general prominent streaks in sequence data. ACM Transactions on Knowledge Discovery from Data 8, 2, Article 9 (June 2014), 37 pages. DOI:https:
//doi.org/10.1145/2601439
Received June 2017; revised August 2018; accepted September 2018