東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁 到查詢結果 [ author_sort:"li, zheng." ]

切換: 標籤 | MARC模式 | ISBD

Pattern and event matching over nois...

Li, Zheng.

FindBook

Google Book

Amazon

博客來

Pattern and event matching over noisy data sequences.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Pattern and event matching over noisy data sequences./
作者:	Li, Zheng.
面頁冊數:	145 p.
附註:	Source: Dissertation Abstracts International, Volume: 76-12(E), Section: B.
Contained By:	Dissertation Abstracts International76-12B(E).
標題:	Computer science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3663998
ISBN:	9781339093031

Pattern and event matching over noisy data sequences.
Li, Zheng.

Pattern and event matching over noisy data sequences. - 145 p.

Source: Dissertation Abstracts International, Volume: 76-12(E), Section: B.

Thesis (Ph.D.)--University of Massachusetts Lowell, 2015.

Sequence data is prevalent in life. Examples include sensor readings, system logs, search engine histories, biological sequences, ECG signals, and smartphone location traces, etc. Such data is usually uncertain/noisy in nature. The noise can be errors in data value, out-of-ordered arrival, missing data and so on. Pattern matching on these streams has diverse applications such as complex event monitoring, medical condition detection, and business intelligence. We study different pattern matching problems under both independent and correlated error model. Under the independent error model, we assume each data value from the raw data have independent error. We first learn a probabilistic sequence from raw data using statistical methods. Then we study (1) substring matching on probabilistic sequences, (2) windowed subsequence matching on probabilistic sequences. Under the correlated error model, we study (3) extended regular expression matching on correlated probabilistic sequences. It can be shown that a former problem is always a special case of a latter problem in the above list. Hence, algorithms devised for a latter problem are more powerful/general, but there is an inherent tradeoff between power and efficiency. Further, we expand this important line of work to handle order error and missing data, we propose order-approximate and parallel-event matching which is needed in many modern applications. Our approaches are supported by novel theoretical analyses as well as a systematic experimental evaluation using both real-world and synthetic datasets.

ISBN: 9781339093031Subjects--Topical Terms:

523869
Computer science.

Pattern and event matching over noisy data sequences.
LDR:02432nmm a2200277 4500 001 2115883
005 20170417071252.5
008 180830s2015 ||||||||||||||||| ||eng d
020 $a 9781339093031
035 $a (MiAaPQ)AAI3663998
035 $a AAI3663998
040 $a MiAaPQ $c MiAaPQ
100 1 $a Li, Zheng. $3 1019154
245 1 0 $a Pattern and event matching over noisy data sequences.
300 $a 145 p.
500 $a Source: Dissertation Abstracts International, Volume: 76-12(E), Section: B.
500 $a Adviser: Tingjian Ge.
502 $a Thesis (Ph.D.)--University of Massachusetts Lowell, 2015.
520 $a Sequence data is prevalent in life. Examples include sensor readings, system logs, search engine histories, biological sequences, ECG signals, and smartphone location traces, etc. Such data is usually uncertain/noisy in nature. The noise can be errors in data value, out-of-ordered arrival, missing data and so on. Pattern matching on these streams has diverse applications such as complex event monitoring, medical condition detection, and business intelligence. We study different pattern matching problems under both independent and correlated error model. Under the independent error model, we assume each data value from the raw data have independent error. We first learn a probabilistic sequence from raw data using statistical methods. Then we study (1) substring matching on probabilistic sequences, (2) windowed subsequence matching on probabilistic sequences. Under the correlated error model, we study (3) extended regular expression matching on correlated probabilistic sequences. It can be shown that a former problem is always a special case of a latter problem in the above list. Hence, algorithms devised for a latter problem are more powerful/general, but there is an inherent tradeoff between power and efficiency. Further, we expand this important line of work to handle order error and missing data, we propose order-approximate and parallel-event matching which is needed in many modern applications. Our approaches are supported by novel theoretical analyses as well as a systematic experimental evaluation using both real-world and synthetic datasets.
590 $a School code: 0111.
650 4 $a Computer science. $3 523869
650 4 $a Computer engineering. $3 621879
690 $a 0984
690 $a 0464
710 2 $a University of Massachusetts Lowell. $3 1017839
773 0 $t Dissertation Abstracts International $g 76-12B(E).
790 $a 0111
791 $a Ph.D.
792 $a 2015
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3663998