Rekomendasi Topik Artikel Ilmiah Dengan Metode Collaborative Dan Content-Based Filtering Terhadap Profil Keaktifan Penulis.

Pangestu, Muhammad Farras (2022) Rekomendasi Topik Artikel Ilmiah Dengan Metode Collaborative Dan Content-Based Filtering Terhadap Profil Keaktifan Penulis. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05111840000134-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (2MB)

Abstract

Tahap pemilihan sebuah topik dalam proses pembuatan artikel ilmiah merupakan tahapan yang sangat penting. Topik dapat digunakan oleh seorang penulis sebagai landasan penulisan untuk menyampaikan isi dari artikel ilmiah. Berdasarkan riwayat pengerjaan suatu topik dari seorang penulis, dapat dilakukan pembuatan sebuah profil penulis. Pada Tugas Akhir ini dilakukan sebuah perekomendasian topik berdasarkan data profil kekatifan penulis. Metode rekomendasi yang digunakan adalah Collaborative Filtering (CF) dan Content-Based Filtering (CB). Pada metode CF dilakukan klasifikasi penulis berdasarkan Profil Keaktifan Penulis dengan menggunakan metode K-Nearest Neighbor (K-NN), Decision Tree (DT), dan Naïve Bayes (NB). Metode CB akan coba melakukan rekomendasi topik berdasarkan pencarian topik-topik yang memiliki kemiripan dengan topik favorit seorang penulis. Perhitungan kemiripan topik dilakukan berdasarkan kata kunci dari masing-masing topik. Metrik perhitungan kemiripan yang digunakan pada proses tersebut adalah Cosine Similarity, Jaccard Similarity, dan Euclidean Distance. Metode evaluasi yang dilakukan adalah dengan menghitung nilai precision, recall, dan f1-score terhadap kesesuaian hasil rekomendasi terhadap data publikasi artikel ilmiah aktual seorang penulis. Pada metode CF didapat hasil terbaik f1-score sebesar 83,44% berdasarkan hasil klasifikasi penulis dari metode K-NN. Hasil evaluasi dari metode CB didapatkan nilai f1-score 62,92% berdasarkan hasil perhitungan kemiripan topik dengan metrik Jaccard Similarity dan Cosine Similarity.
==============================================================================================================================
The stage of selecting a topic in the process of creating a scientific article is a very important stage. The topic can be used by an author as a basis for writing to convey the content of a scientific article. Based on the history of working on a topic from an author, it is possible to create an author profile. In this Final Project, a topic recommendation is carried out based on the Author's Activeness Profile data. The recommendation methods used are Collaborative Filtering (CF) and Content-Based Filtering (CB). In the CF method, the classification of authors was carried out based on the Author's Activeness Profile using the K-Nearest Neighbor (K-NN), Decision Tree (DT), and Naïve Bayes (NB) methods. The CB method will try to make topic recommendations based on the search for those topics that have similarities with the favorite topic of an author. The calculation of the similarity of topics is carried out based on the keywords of each topic. The similarity calculation metrics used in the process are Cosine Similarity, Jaccard Similarity, and Euclidean Distance. The evaluation method carried out is to calculate the precision, recall, and f1-score values for the suitability of the recommendation results to the data of the publication of an author's actual scientific article. In the CF method, the best result of f1-score was obtained at 83.44% based on the results of the author classification of the K-NN method. The evaluation results of the CB method obtained an f1-score value of 62.92% based on the results of calculating the similarity of the topic with the Jaccard Similarity and Cosine Similarity metrics.

Item Type:	Thesis (Other)
Additional Information:	RSIf 006.312 Pan r-1 2022
Uncontrolled Keywords:	Collaborative Filtering, Content-Based Filtering, Penulis, Topik, Rekomendasi. Author, Collaboration Filtering, Content-Based Filtering, Topic, Recommendation.
Subjects:	Q Science > QA Mathematics > QA76.9.I58 Recommender systems (Information filtering)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	26 May 2026 04:19
Last Modified:	26 May 2026 04:19
URI:	http://repository.its.ac.id/id/eprint/133436

Actions (login required)

View Item