MARC details
000 -LEADER |
fixed length control field |
02345nam a22002893a 4500 |
001 - CONTROL NUMBER |
control field |
UPMIN-00004810112 |
003 - CONTROL NUMBER IDENTIFIER |
control field |
UPMIN |
005 - DATE AND TIME OF LATEST TRANSACTION |
control field |
20230201170028.0 |
008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION |
fixed length control field |
230201b |||||||| |||| 00| 0 eng d |
040 ## - CATALOGING SOURCE |
Original cataloging agency |
DLC |
Transcribing agency |
UPMin |
Modifying agency |
upmin |
041 ## - LANGUAGE CODE |
Language code of text/sound track or separate title |
eng |
090 #0 - LOCALLY ASSIGNED LC-TYPE CALL NUMBER (OCLC); LOCAL CALL NUMBER (RLIN) |
Classification number (OCLC) (R) ; Classification number, CALL (RLIN) (NR) |
LG993.5 2010 |
Local cutter number (OCLC) ; Book number/undivided call number, CALL (RLIN) |
A64 M34 |
100 ## - MAIN ENTRY--PERSONAL NAME |
Personal name |
Madarang, Jennelle Rizza M. |
9 (RLIN) |
2021 |
245 ## - TITLE STATEMENT |
Title |
Modified K-mean clustering algorithm for fixed numeric and categorical data sets with missing values / |
Statement of responsibility, etc. |
Jennelle Rizza M. Madarang |
260 ## - PUBLICATION, DISTRIBUTION, ETC. |
Date of publication, distribution, etc. |
2010 |
300 ## - PHYSICAL DESCRIPTION |
Extent |
65 leaves. |
502 ## - DISSERTATION NOTE |
Dissertation note |
Thesis (BS Applied Mathematics) -- University of the Philippines Mindanao, 2010 |
520 3# - SUMMARY, ETC. |
Summary, etc. |
Clustering is a data mining technique that aims to organize a given set of objects into groups or clusters such that objects within the same cluster are more similar to each other than to data objects in other clusters. However, most of the clustering algorithms deal with complete and with either numeric or categorical data sets only, but not mixed. Ahmad and Dey (2007) proposed an algorithm for clustering complete mixed data sets. In order to deal with incomplete data sets or missing values, modification of the proposed algorithm of Ahmad and Dey (2007) was done. The modification combined two techniques of handling missing values which are available case analysis which uses the available information left on the data set, and the adaptive imputation which imputes missing data during the clustering stage. The performance of the modified algorithm was tested in two data sets, small and large, and was compared to other existing methods namely, case deletion, mean and mode imputation, and kNN imputation using the Adjusted Ran Index, modified algorithm produced fair quality of resulting clusters in the small data set. It was competitive with regards to K-mean after mean and mode imputation and K-mean after kNN imputation. However, the quality of the resulting clusters on large data set is very poor on all methods. It seemed that as the size of the data set becomes bigger the modified K-mean algorithm performed worse. |
650 17 - SUBJECT ADDED ENTRY--TOPICAL TERM |
Topical term or geographic name entry element |
Clustering |
9 (RLIN) |
366 |
650 17 - SUBJECT ADDED ENTRY--TOPICAL TERM |
Topical term or geographic name entry element |
K-mean algorithm |
9 (RLIN) |
2022 |
650 17 - SUBJECT ADDED ENTRY--TOPICAL TERM |
Topical term or geographic name entry element |
Missing values |
9 (RLIN) |
990 |
650 17 - SUBJECT ADDED ENTRY--TOPICAL TERM |
Topical term or geographic name entry element |
Mixed numeric and categorical data |
9 (RLIN) |
2023 |
658 ## - INDEX TERM--CURRICULUM OBJECTIVE |
Main curriculum objective |
Undergraduate Thesis |
Curriculum code |
AMAT200 |
905 ## - LOCAL DATA ELEMENT E, LDE (RLIN) |
a |
Fi |
905 ## - LOCAL DATA ELEMENT E, LDE (RLIN) |
a |
UP |
942 ## - ADDED ENTRY ELEMENTS (KOHA) |
Source of classification or shelving scheme |
Library of Congress Classification |
Koha item type |
Thesis |