Page 109 - profile-ok

研究人員 ｜ Research Faculty

DEGREES RECEIVED: ACADEMIC HONORS AND AWARDS:
葉彌妍 Mi-Yen Yeh ● Ph.D., EE, National Taiwan University, Taiwan (2002/09 - 2009/01) ƒ Distinguished Postdoctoral Fellowship, Academia Sinica, Taiwan
● B.S., EE, National Taiwan University, Taiwan (1998/09 - 2002/06) (2009)

PROFESSIONAL EXPERIENCE:
助研究員 Assistant Research Fellow ● Postdoctoral Fellow, CITI, Academia Sinica, Taiwan (2009/02 -
Ph.D., Electrical Engineering, National Taiwan University 2009/09)
● Visiting Scientist, IBM T. J. Watson Research Center, NY, USA (2007/09
Tel: +886-2-2788-3799 ext. 1412 - 2008/07)
Fax: +886-2-2782-4814
Email: miyen@iis.sinica.edu.tw
http://www.iis.sinica.edu.tw/pages/miyen

代表著作 Publications

1. Bi-Ru Dai, Jen-Wei Huang, Mi-Yen Yeh, and Ming-Syan Chen,
”Clustering on Demand for Multiple Data Streams”, In Proceedings
研究簡介 Research Description of the IEEE 4th International Conference on Data Mining (ICDM),
November 1-4, 2004.
我的研究興趣主要是在資料探勘和資料庫方面的相關議 My research interests lie in the area of data mining and databases. Particularly, I have 2. Mi-Yen Yeh, Bi-Ru Dai and Ming-Syan Chen, “COMET: Event-Driv-
題。過去幾年的研究主要是針對多重數值資料流環境， been focusing on topics in a multiple-data-stream environment, including stream en Clustering over Multiple Evolving Streams,” In Proceedings of the
10th Pacific-Asia Conf. on Knowledge Discovery and Data Mining
探討資料摘要、資料叢集、以及相似性查詢處理。資料 summarization, clustering, and similarity query processing. The most important (PAKDD), April 9-12, 2006.
流的特性是持續收集、快速匯入、且大量累積，該如何 characteristics of data streams are continuously arriving, of huge volumes, and with 3. Bi-Ru Dai, Jen-Wei Huang, Mi-Yen Yeh, and Ming-Syan Chen,
設計即時、有限次數檢視、以及限量資料儲存的演算法 fast coming speeds. To accommodate these properties of data streams with limited “Adaptive Clustering for Multiple Evolving Streams,” IEEE Transac-
便成為重要的議題。 computing resources and bounded storage, it is important to design approximate tions on Knowledge and Data Engineering, Vol. 18, No. 9, pp. 1166-
but real-time algorithms. 1180, September 2006.
另外，在多數應用中，資料流的取得常是獨立且分佈在
不同的位置。對於相似性查詢，若先將所有資料集中再 In addition, I am interested in similarity query processing among distributed 4. Mi-Yen Yeh, Bi-Ru Dai, and Ming-Syan Chen, “Clustering over Mul-
tiple Evolving Streams by Events and Correlations,” IEEE Transac-
處理並非一個有效率的方法。因此，探討如何直接在分 streams. In many real-world applications, data streams are collected independently tions on Knowledge and Data Engineering, Vol. 19, No. 10, pp. 1349-
散式環境底下，有效地節省頻寬使用來傳送資料流摘要 in a decentralized manner. For example, streams of climate measurements such as 1362, October 2007.
以獲得查詢的答案，是另一個值得研究的問題。 temperatures are collected from observation stations located over a wide area. It 5. Mi-Yen Yeh, Kun-Lung Wu, Philip S. Yu, and Ming-Syan Chen,
is inefficient to gather all of the distributed streams to a central site before doing “LEEWAVE: Level-Wise Distribution of Wavelet Coefficients for
最後，如何處理含有不確定性的資料，亦是一個重要研 any query processing especially when the available network bandwidth is limited. Processing kNN Queries over Distributed Streams,” In Proceedings of
究議題。受到儀器誤差影響，或是為保護隱私而受到人 Hence, there is a need to develop a bandwidth-efficient approach to processing the 34th International Conference on Very Large Data Bases (VLDB),
為刻意模糊，我們所獲得的資料並非就是所知的定值， queries among distributed streams. August 24-30, 2008.
而應該是一個可能值的機率分佈。由於資料的不確定 6. Mi-Yen Yeh, Kun-Lung Wu, Philip S. Yu, and Ming-Syan Chen,
性，傳統的資料模型和相似度定義將不再適用。因此需 Last but not least, how to do similarity query processing on data with uncertainty “PROUD: A Probabilistic Approach to Processing Similarity Queries
要藉由機率和統計理論輔助，建立不確定性物體之間距 is also a topic that I concern about. Uncertainty of data comes from various sources over Uncertain Data Streams,” In Proceedings of the 12th Internation-
離的模型，定義新的資料相似性查詢模式並設計對應的 such as noise interference of equipments or man-made data perturbation for pri- al Conference on Extending Database Technology (EDBT), March
23-26, 2009.
處理方法。 vacy preservation. In contrast to data with certainty, the data are now regarded as
random variables with probability distributions. Hence, new probabilistic models 7. Su-Chen Lin, Mi-Yen Yeh, and Ming-Syan Chen, “Subsequence
and methods for similarity queries among uncertain data should be designed. Matching of Stream Synopses under the Time Warping Distance,” In
Proceedings of the Pacific-Asia Conference on Knowledge Discovery
and Data Mining (PAKDD), June 21-24, 2010.

108 109