![](bga.png)
formed.
Experiments performed over data from both syn-
thetic and real applications showed that our technique
indeed perform in average 3 times faster them the
state of the art approach (Random Projection), includ-
ing the best methods previously available. It also re-
quires less memory, and the experiments revealed that
it requires up to 10 times less memory than the com-
petitor methods. We presented the results to special-
ists in the field (meteorologists), that confirmed that
the results are indeed correct and useful for their day-
to-day activities to process climate data. For future
works, we intend to explore data from larger regions,
as the whole Brazil and South America. We also in-
tend to explore data from different sensors in order to
evaluate improvements that may be needed.
ACKNOWLEDGEMENTS
The authors are grateful for the financial support
granted by FAPESP, CNPq, CAPES, SticAmsud, Em-
brapa Agricultural Informatics, Cepagri/Unicamp and
Agritempo for data.
REFERENCES
Catalano, J., Armstrong, T., and Oates, T. (2006). Dis-
covering Patterns in Real-Valued Time Series. In
F
¨
urnkranz, J., Scheffer, T., and Spiliopoulou, M., ed-
itors, Knowledge Discovery in Databases: PKDD
2006, volume 4213 of Lecture Notes in Computer Sci-
ence, pages 462–469. Springer Berlin Heidelberg.
Chiu, B., Keogh, E., and Lonardi, S. (2003). Probabilis-
tic discovery of time series motifs. In Proceedings
of the ninth ACM SIGKDD international conference
on Knowledge discovery and data mining, KDD ’03,
pages 493–498, New York, NY, USA. ACM.
Faloutsos, C., Ranganathan, M., Manolopoulos, Y., and
Manolopoulos, Y. (1994). Fast subsequence match-
ing in time-series databases. In Proceedings of the
ACM SIGMOD International Conference on Manage-
ment of Data, pages 419–429, Minneapolis, USA.
Goldin, D. Q., Kanellakis, P. C., and Kanellakis, P. C.
(1995). On similarity queries for time-series data:
Constraint specification and implementation. In Pro-
ceedings of the 1st International Conference on Prin-
ciples and Practice of Constraint Programming, pages
137–153, Cassis, France.
Keogh, E., Chakrabarti, K., Pazzani, M., and Mehrotra, S.
(2001). Dimensionality reduction for fast similarity
search in large time series databases. Knowledge and
Information Systems, 3:263–286.
Keogh, E. and Kasetty, S. (2003). On the need for time
series data mining benchmarks: a survey and empir-
ical demonstration. In Data Mining and Knowledge
Discovery, volume 7, pages 349–371. Springer.
Keogh, E., Lin, J., Lee, S.-H., and Herle, H. (2007). Find-
ing the most unusual time series subsequence: algo-
rithms and applications. Knowledge and Information
Systems, 11(1):1–27.
Li, Y. and Lin, J. (2010). Approximate variable-length
time series motif discovery using grammar inference.
In Proceedings of the Tenth International Workshop
on Multimedia Data Mining, MDMKDD ’10, pages
10:1—-10:9, New York, NY, USA. ACM.
Li, Y., Lin, J., and Oates, T. (2012). Visualizing Variable-
Length Time Series Motifs. In SDM, pages 895–906.
SIAM / Omnipress.
Lin, J., Keogh, E., Lonardi, S., and Chiu, B. (2003). A sym-
bolic representation of time series, with implications
for streaming algorithms. In Proceedings of the 8th
ACM SIGMOD workshop on Research issues in data
mining and knowledge discovery, DMKD ’03, pages
2–11, New York, NY, USA. ACM.
Lin, J., Keogh, E., Patel, P., and Lonardi, S. (2002). Finding
motifs in time series. In In the 2nd Workshop on Tem-
poral Data Mining, at the 8th ACM SIGKDD Interna-
tional Conference on Knowledge Discovery and Data
Mining, International Conference on Knowledge Dis-
covery and Data Mining, Edmonton, Alberta, Canada.
ACM.
Lin, J., Keogh, E. J., Wei, L., and Lonardi, S. (2007). Ex-
periencing sax: a novel symbolic representation of
time series. Data Mining and Knowledge Discovery,
15:107–144.
Mohammad, Y. and Nishida, T. (2009). Constrained Motif
Discovery in Time Series. New Generation Comput-
ing, 27(4):319–346.
Rouse, J. W., Haas, R. H., Schell, J. A., and Deering, D. W.
(1973). Monitoring vegetation systems in the great
plains with ERTS. In Proceedings of the Third ERTS
Symposium, pages 309–317, Washington, DC, USA.
Udechukwu, A., Barker, K., and Alhajj, R. (2004). Discov-
ering all frequent trends in time series. In Proceedings
of the winter international synposium on Information
and communication technologies, WISICT ’04, pages
1–6. Trinity College Dublin.
Wang, L., Chng, E. S., and Li, H. (2010). A tree-
construction search approach for multivariate time se-
ries motifs discovery. Pattern Recognition Letters,
31(9):869–875.
Yankov, D., Keogh, E., Medina, J., Chiu, B., and Zordan,
V. (2007). Detecting time series motifs under uniform
scaling. In Proceedings of the 13th ACM SIGKDD
international conference on Knowledge discovery and
data mining, KDD ’07, pages 844–853, New York,
NY, USA. ACM.
TrieMotif-ANewandEfficientMethodtoMineFrequentK-MotifsfromLargeTimeSeries
69