Analisis Rekomendasi Menggunakan Single Linkage Clustering Dan K-Modes Clustering Dalam Pendekatan Hybrid Filtering

Hidayatullah, Anadia Rahmat Syihab (2019) Analisis Rekomendasi Menggunakan Single Linkage Clustering Dan K-Modes Clustering Dalam Pendekatan Hybrid Filtering. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06211540000001-Undergraduate_Theses.pdf]
Preview
Text
06211540000001-Undergraduate_Theses.pdf

Download (2MB) | Preview

Abstract

Perkembangan teknologi telah membawa perubahan dalam bidang kehidupan manusia. Hingga pada saat era digital kini semua media informasi dapat diakses melalui internet. Salah satu media yang memanfaatkan teknologi internet dalam memperoleh keuntungan adalah dunia film dengan cara menyajikan film secara daring. Sejak awal perkembangannya hingga saat ini telah tercatat 3.361.741 judul film yang telah dikeluarkan industri perfilman. Sekian banyak film membawa dampak bagi user dan pemilik situs film daring terutama dalam hal efisiensi tampilan yang film yang relevan bagi user-user film. Oleh karena itu munculah pendekatan yang dapat memberikan bantuan bagi user dalam menentukan keputusan yakni menggunakan kombinasi sistem rekomendasi yakni teknik sistem rekomendasi Hybrid Filtering yaitu menggabungkan teknik Demographic Filtering (Single Linkage Clustering dan K-Modes Clustering) dan Collaborative Filtering. Hasil dari penelitian diperoleh jumlah kelompok user berdasarkan kemiripan karakteristik yang sama sejumlah 34 jenis klaster. Rekomendasi yang didapatkan sejumlah 10 film yang relevan melalui proses preferensi user lain dan prediksi rating.
===============================================================================================================================
Technological developments have brought changes in the field of human living. Until right now in this digital era, all media information can be accessed via the internet. One of the media that uses internet tech-nology to gain profits is the world of movie by presenting movies online. Since the beginning of its development to date there have been 3,361,741 movie titles released by the movie industry. Those many movies have an impact on the users and owners of online movie sites, especially in terms of the efficiency of the appearance of the movies that are relevant to the users of the movie. Therefore, an approach that can provide assistance to users in determining decisions is using a combination of recommendation systems, it called a Hybrid Filtering recommendation system technique that combines Demographic Filtering techniques (Single Linkage Clus-tering and K-Modes Clustering) and Collaborative Filtering. The results of the study obtained the number of user groups based on similar charac-teristics of 34 types of clusters. The recommendations obtained are 10 relevant movies through other user preference processes and rating predictions.

Item Type: Thesis (Other)
Additional Information: RSSt 519.535 Hid a-1 2019
Uncontrolled Keywords: Sistem Rekomendasi, Film, Single Linkage Clustering, K-Modes Clustering, Hybrid Filtering
Subjects: H Social Sciences > HA Statistics
H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics
H Social Sciences > HE Transportation and Communications
H Social Sciences > HF Commerce > HF54.54 Electronic information resources. Digital libraries
Q Science > QA Mathematics > QA278.55 Cluster analysis
Divisions: Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: ANADIA RAHMAT SYIHAB HIDAYATULLAH
Date Deposited: 17 Mar 2023 07:31
Last Modified: 17 Mar 2023 07:52
URI: http://repository.its.ac.id/id/eprint/63697

Actions (login required)

View Item View Item