國立虎尾科技大學 |

Automated Classification to Improve the Efficiency of Weeding Library Collections.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Automated Classification to Improve the Efficiency of Weeding Library Collections./
作者:	Wagstaff, Kiri L.
面頁冊數:	1 online resource (68 pages)
附註:	Source: Masters Abstracts International, Volume: 56-05.
Contained By:	Masters Abstracts International56-05(E).
標題:	Library science. -
電子資源:	click for full text (PQDT)
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. - 1 online resource (68 pages)

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

Thesis (M.L.I.S.)

Includes bibliographical references

Studies have shown that library weeding (the selective removal of unused, worn, outdated, or irrelevant items) benefits patrons and increases circulation rates.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weeding projects could use the same approach to train a model to quickly identify the candidates most likely to be withdrawn.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355068825Subjects--Topical Terms:

561163
Library science.
Index Terms--Genre/Form:

554714
Electronic books.

Automated Classification to Improve the Efficiency of Weeding Library Collections.
LDR:02583ntm a2200349Ki 4500 001 911569
005 20180529094436.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9780355068825
035 $a (MiAaPQ)AAI10286764
035 $a (MiAaPQ)sjsu:11132
035 $a AAI10286764
040 $a MiAaPQ $b eng $c MiAaPQ
099 $a TUL $f hyy $c available through World Wide Web
100 1 $a Wagstaff, Kiri L. $3 1183480
245 1 0 $a Automated Classification to Improve the Efficiency of Weeding Library Collections.
264 0 $c 2017
300 $a 1 online resource (68 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 56-05.
500 $a Adviser: Geoffrey Z. Liu.
502 $a Thesis (M.L.I.S.) $c San Jose State University $d 2017.
504 $a Includes bibliographical references
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.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Library science. $3 561163
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0399
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a San Jose State University. $b Library and Information Science. $3 1183481
773 0 $t Masters Abstracts International $g 56-05(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10286764 $z click for full text (PQDT)