東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Automated Classification to Improve ...

Wagstaff, Kiri L.

Linked to FindBook

Google Book

Amazon

博客來

Automated Classification to Improve the Efficiency of Weeding Library Collections.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Automated Classification to Improve the Efficiency of Weeding Library Collections./
Author:	Wagstaff, Kiri L.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2017,
Description:	68 p.
Notes:	Source: Masters Abstracts International, Volume: 56-05.
Contained By:	Masters Abstracts International56-05(E).
Subject:	Library science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10286764
ISBN:	9780355068825

Automated Classification to Improve the Efficiency of Weeding Library Collections.
Wagstaff, Kiri L.

Automated Classification to Improve the Efficiency of Weeding Library Collections. - Ann Arbor : ProQuest Dissertations & Theses, 2017 - 68 p.

Source: Masters Abstracts International, Volume: 56-05.

Thesis (M.L.I.S.)--San Jose State University, 2017.

Studies have shown that library weeding (the selective removal of unused, worn, outdated, or irrelevant items) benefits patrons and increases circulation rates. However, the time required to review the collection and make weeding decisions presents a formidable obstacle. In this study, we empirically evaluated methods for automatically classifying weeding candidates. A data set containing 80,346 items from a large-scale academic library weeding project by Wesleyan University from 2011 to 2014 was used to train six machine learning classifiers to predict "Keep" or "Weed" for each candidate. We found statistically significant agreement (p = 0.001) between classifier predictions and librarian judgments for all classifier types. The naive Bayes and linear support vector machine classifiers had the highest recall (fraction of items weeded by librarians that were identified by the algorithm), while the k-nearest-neighbor classifier had the highest precision (fraction of recommended candidates that librarians had chosen to weed). The most relevant variables were found to be librarian and faculty votes for retention, item age, and the presence of copies in other libraries. Future weeding projects could use the same approach to train a model to quickly identify the candidates most likely to be withdrawn.

ISBN: 9780355068825Subjects--Topical Terms:

539284
Library science.

Automated Classification to Improve the Efficiency of Weeding Library Collections.
LDR:02249nmm a2200301 4500 001 2111700
005 20180529094436.5
008 180626s2017 ||||||||||||||||| ||eng d
020 $a 9780355068825
035 $a (MiAaPQ)AAI10286764
035 $a (MiAaPQ)sjsu:11132
035 $a AAI10286764
040 $a MiAaPQ $c MiAaPQ
100 1 $a Wagstaff, Kiri L. $3 3265838
245 1 0 $a Automated Classification to Improve the Efficiency of Weeding Library Collections.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2017
300 $a 68 p.
500 $a Source: Masters Abstracts International, Volume: 56-05.
500 $a Adviser: Geoffrey Z. Liu.
502 $a Thesis (M.L.I.S.)--San Jose State University, 2017.
520 $a Studies have shown that library weeding (the selective removal of unused, worn, outdated, or irrelevant items) benefits patrons and increases circulation rates. However, the time required to review the collection and make weeding decisions presents a formidable obstacle. In this study, we empirically evaluated methods for automatically classifying weeding candidates. A data set containing 80,346 items from a large-scale academic library weeding project by Wesleyan University from 2011 to 2014 was used to train six machine learning classifiers to predict "Keep" or "Weed" for each candidate. We found statistically significant agreement (p = 0.001) between classifier predictions and librarian judgments for all classifier types. The naive Bayes and linear support vector machine classifiers had the highest recall (fraction of items weeded by librarians that were identified by the algorithm), while the k-nearest-neighbor classifier had the highest precision (fraction of recommended candidates that librarians had chosen to weed). The most relevant variables were found to be librarian and faculty votes for retention, item age, and the presence of copies in other libraries. Future weeding projects could use the same approach to train a model to quickly identify the candidates most likely to be withdrawn.
590 $a School code: 6265.
650 4 $a Library science. $3 539284
650 4 $a Artificial intelligence. $3 516317
690 $a 0399
690 $a 0800
710 2 0 $a San Jose State University. $b Library and Information Science. $3 2100156
773 0 $t Masters Abstracts International $g 56-05(E).
790 $a 6265
791 $a M.L.I.S.
792 $a 2017
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10286764