TEMU KEMBALI INFORMASI PADA PENCARIAN JURNAL SKRIPSI MENGGUNAKAN METODE SINGLE PASS CLUSTERING
Total View This Week0
Total View This Week0
Institusion
Universitas Muhammadiyah Gresik
Author
SAFITRI, RIMA NOER
Subject
Informatics Engineering
Datestamp
2020-12-11 08:38:49
Abstract :
Journal thesis is often used as a source of reference in making a study. The greater the number of thesis collection owned by the speed and accuracy of information is very important for the acquisition of information seekers. This study seeks to apply information retrieval with single pass clustering methods for finding results in the form of an abstract thesis that has relevance to the user query input. System testing conducted empirical tests with 13 queries and threshold input from 0-0.25. The data used for the test system is a collection of abstracts of theses documents Informatics Engineering Program UMG years 2007-2012 were taken 73 documents from digital library UMG abstract document. The selection of the threshold value will determine the performance of the system. Based on the test results obtained are quite ideal threshold value is 0:15 with precission value 74.4505%, recall value reached 68.0769% and 96.1012 % accuracy achieved.
Journal thesis is often used as a source of reference in making a study. The greater the number of thesis collection owned by the speed and accuracy of information is very important for the acquisition of information seekers. This study seeks to apply information retrieval with single pass clustering methods for finding results in the form of an abstract thesis that has relevance to the user query input. System testing conducted empirical tests with 13 queries and threshold input from 0-0.25. The data used for the test system is a collection of abstracts of theses documents Informatics Engineering Program UMG years 2007-2012 were taken 73 documents from digital library UMG abstract document. The selection of the threshold value will determine the performance of the system. Based on the test results obtained are quite ideal threshold value is 0:15 with precission value 74.4505%, recall value reached 68.0769% and 96.1012 % accuracy achieved.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
