Topic Discovery Pada Dokumen Abstrak Jurnal Penelitian Di Science Direct Menggunakan Association Rule

Fatchur Roji, Mochammad Farros (2019) Topic Discovery Pada Dokumen Abstrak Jurnal Penelitian Di Science Direct Menggunakan Association Rule. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211540000098-Undergraduate_Theses.pdf Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Jurnal memiliki peranan penting dalam meningkatkan pemahaman mengenai ilmu berdasarkan review dari ilmuwan. Jurnal yang telah dibuat dalam bentuk digitalisasi memiliki istilah lain yaitu file atau soft copy dengan memanfaatkan teknologi informasi dan komunikasi, yang saat ini menjadi salah satu koleksi perpustakaan digital. Tujuan penelitian ini adalah untuk menemukan kandidat topik menggunakan association rule dan melakukan efisiensi menggunakan closed frequent itemset ditambah remove subset serta korelasi berdasarkan tahun terbit jurnal tersebut. Data yang di gunakan dalam penelitian ini berasal dari ScienceDirect. Dokumen abstrak dari ScienceDirect tersebut nantinya akan dilakukan pre processing terlebih dahulu, kemudian di lanjutkan dengan association rule dan pearson correlation setelahnya. Pada association rule term kata jika menggunakan min support 2 % maka di dapatkan frequent itemset sebanyak 72, closed frequent itemset sebanyak 55, dan remove subset sebanyak 41 itemset. Kemudian saat di lakukan analisis korelasi pada itemset remove subset. Di dapatkan bayesian,model yaitu itemset yang paling banyak memiliki hubungan. Kemudian pada topic community dengan cfinder terbagi menjadi dua komunitas dan terdapat irisan sebanyak 6 itemset.
=================================================================================================================================
Journals have an important role in increasing understanding of science based on reviews from scientists. Characteristics of journals such as updates are related to theory compared to books, a more concise discussion, as a reference to real-world alternatives, applications and implementations. Journals that have been made in the form of digitalization have other terms, namely file or soft copy by utilizing information and communication technology, which is currently one of the digital library collections. The data used is from ScienceDirect. ScienceDirect is a database that contains quality full-text documents that have been reviewed by Elsevier. The abstract document from Sciencedirect will be pre-processed first. Then proceed with the association rule and the pearson correlation afterwards. In the association rule term, if you use min support of 2%, you get 72 frequent itemset, 55 frequent frequent itemset, and 41 itemset remove subset. Then when doing a correlation analysis on itemset remove subset. In getting bayesian,model that are itemset that have the most relationships. Then on the topic community with cfinder divided into two communities and there are 6 itemset slices.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Association Rule, E-Jurnal, Pearson Correlation, Pre processing, ScienceDirect
Subjects:	Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions:	Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Roji Mochammad Farros Fatchur
Date Deposited:	23 Jul 2026 03:10
Last Modified:	23 Jul 2026 03:10
URI:	http://repository.its.ac.id/id/eprint/67572

Actions (login required)

View Item