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

切換: 標籤 | MARC模式 | ISBD

PG-means: Learning the number of clu...

Feng, Yu.

FindBook

Google Book

Amazon

博客來

PG-means: Learning the number of clusters in data.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	PG-means: Learning the number of clusters in data./
作者:	Feng, Yu.
面頁冊數:	61 p.
附註:	Source: Masters Abstracts International, Volume: 45-03, page: 1527.
Contained By:	Masters Abstracts International45-03.
標題:	Artificial Intelligence. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1440574

PG-means: Learning the number of clusters in data.
Feng, Yu.

PG-means: Learning the number of clusters in data. - 61 p.

Source: Masters Abstracts International, Volume: 45-03, page: 1527.

Thesis (M.S.)--Baylor University, 2007.

We present a novel algorithm called PG-means in this thesis. This algorithm is able to determine the number of clusters in a classical Gaussian mixture model automatically. PG-means uses efficient statistical hypothesis tests on one-dimensional projections of the data and model to determine if the examples are well represented by the model. In so doing, we apply a statistical test to the entire model at once, not just on a per-cluster basis. We show that this method works well in difficult cases such as overlapping clusters, eccentric clusters and high dimensional clusters. PG-means also works well on non-Gaussian clusters and many true clusters. Further, the new approach provides a much more stable estimate of the number of clusters than current methods.Subjects--Topical Terms:

769149
Artificial Intelligence.

PG-means: Learning the number of clusters in data.
LDR:01572nmm 2200265 4500 001 1834668
005 20071127115007.5
008 130610s2007 eng d
035 $a (UMI)AAI1440574
035 $a AAI1440574
040 $a UMI $c UMI
100 1 $a Feng, Yu. $3 729715
245 1 0 $a PG-means: Learning the number of clusters in data.
300 $a 61 p.
500 $a Source: Masters Abstracts International, Volume: 45-03, page: 1527.
500 $a Adviser: Gregory J. Hamerly.
502 $a Thesis (M.S.)--Baylor University, 2007.
520 $a We present a novel algorithm called PG-means in this thesis. This algorithm is able to determine the number of clusters in a classical Gaussian mixture model automatically. PG-means uses efficient statistical hypothesis tests on one-dimensional projections of the data and model to determine if the examples are well represented by the model. In so doing, we apply a statistical test to the entire model at once, not just on a per-cluster basis. We show that this method works well in difficult cases such as overlapping clusters, eccentric clusters and high dimensional clusters. PG-means also works well on non-Gaussian clusters and many true clusters. Further, the new approach provides a much more stable estimate of the number of clusters than current methods.
590 $a School code: 0014.
650 4 $a Artificial Intelligence. $3 769149
650 4 $a Computer Science. $3 626642
690 $a 0800
690 $a 0984
710 2 0 $a Baylor University. $3 771397
773 0 $t Masters Abstracts International $g 45-03.
790 1 0 $a Hamerly, Gregory J., $e advisor
790 $a 0014
791 $a M.S.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1440574