Pattern Discovery and Change Detection of Online Music Query Streams

  • Hua-Fu Li


In this paper, an efficient stream mining algorithm, called FTP-stream (Frequent Temporal Pattern mining of streams), is proposed to find the frequent temporal patterns over melody sequence streams. In the framework of our proposed algorithm, an effective bit-sequence representation is used to reduce the time and memory needed to slide the windows. The FTP-stream algorithm can calculate the support threshold in only a single pass based on the concept of bit-sequence representation. It takes the advantage of “left” and “and” operations of the representation. Experiments show that the proposed algorithm only scans the music query stream once, and runs significant faster and consumes less memory than existing algorithms, such as SWFI-stream and Moment.


Data Stream Frequent Pattern Pattern Mining Frequent Itemsets Mining Frequent Itemsets 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



The authors thank the reviewers’ precious comments for improving the quality of the paper. We would like to thank Dr. Yun Chi for contributing the source codes of Moment algorithm (MomentFP). The research is supported in part by the National Science Council, Project No. NSC 96-2218-E-424-001-, Taiwan, Republic of China.


  1. 1.
    Agrawal R, Srikant R (1994) Fast algorithms for mining association rules. In: Proc. VLDB, pp 487–499Google Scholar
  2. 2.
    Babcock B, Babu S, Data M, Motwani R, Widom J (2002) Models and issues in data stream systems. In: Proc. PODS, pp 1–16Google Scholar
  3. 3.
    Bakhmutora V, Gusev VU, Titkova TN (1997) The search for adaptations in song melodies. Comput Music J 21(1):58–67. doi:10.2307/3681219CrossRefGoogle Scholar
  4. 4.
    Chang JH, Lee WS (2003) Finding recent frequent itemsets adaptively over online data streams. In: Proc. KDD, pp 487–492Google Scholar
  5. 5.
    Chang JH, Lee WS (2004) A sliding window method for finding recently frequent itemsets over online data streams. J Inform Sci Eng (JISE) 20(4):753–762Google Scholar
  6. 6.
    Chi Y, Wang H, Yu P, Muntz R (2004) MOMENT: maintaining closed frequent itemsets over a stream sliding window. In: Proc. ICDM, pp 59–66Google Scholar
  7. 7.
    Dong G, Han J, Lakshmanan LVS, Pei J, Wang H, Yu PS (2003) Online mining of changes from data streams: research problems and preliminary results. In: Proc. ACM SIGMOD-MPDSGoogle Scholar
  8. 8.
    Gaber MM, Zaslavsky A, Krishnaswamy S (2005) Mining data streams: a review. SIGMOD Rec 34(1): 18–26CrossRefGoogle Scholar
  9. 9.
    Hsu J-L, Liu C-C, Chen ALP (2001) Discovering nontrivial repeating patterns in music data. IEEE Trans Multimed 3(3):311–325. doi:10.1109/6046.944475CrossRefGoogle Scholar
  10. 10.
    Jones GT (1974) Music theory. Harper & Row, New YorkGoogle Scholar
  11. 11.
    Li H-F, Lee S-Y, Shan M-K (2004) Mining frequent closed structures in streaming melody sequences. In: Proc. ICMEGoogle Scholar
  12. 12.
    Li H-F, Lee S-Y, Shan M-K (2005) Online mining maximal frequent structures in continuous landmark melody streams. Pattern Recognit Lett 26(11):1658–1674. doi:10.1016/j.patrec.2005.01.016CrossRefGoogle Scholar
  13. 13.
    Li H-F, Lee S-Y, Shan M-K (2005) Online mining changes of items over continuous append-only and dynamic data streams. J Univers Comput Sci 11(8):1411–1425Google Scholar
  14. 14.
    Shan M-K, Kuo F-F (2003) Music style mining and classification by melody. IEICE Trans Inform Syst E 86-D(4):655–659Google Scholar
  15. 15.
    Yoshitaka A, Ichikawa T (1999) A survey on content-based retrieval for multimedia databases. IEEE Trans Knowl Data Eng 11(1):81–93. doi:10.1109/ 69.755617CrossRefGoogle Scholar
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.

Copyright information

© Springer Science+Business Media, LLC 2009

Authors and Affiliations

  1. 1.Department of Computer ScienceKainan UniversityTaoyuanTaiwan

Personalised recommendations