Pengembangan Sistem Rekomendasi Mobil Bekas Berbasis Content-Based Filtering Menggunakan Data OLX.co.id

Himawan, Farhan (2025) Pengembangan Sistem Rekomendasi Mobil Bekas Berbasis Content-Based Filtering Menggunakan Data OLX.co.id. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5026211125-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Pasar mobil bekas di Indonesia mengalami pertumbuhan signifikan seiring perkembangan teknologi dan meningkatnya penggunaan platform e-commerce seperti OLX.co.id. Banyaknya pilihan mobil bekas dengan atribut berbeda, seperti harga, tahun produksi, dan jenis bahan bakar, seringkali menyulitkan konsumen dalam menemukan kendaraan yang sesuai preferensi. Penelitian ini mengembangkan sistem rekomendasi mobil bekas berbasis Content-Based Filtering (CBF) dengan pendekatan K-Nearest Neighbors (KNN), yang bekerja berdasarkan kemiripan atribut antara input pengguna dan data mobil. Sebelum proses rekomendasi dilakukan, data diklaster menggunakan algoritma K-Means untuk menyaring ruang pencarian berdasarkan kesamaan fitur numerik dan kategorikal. Hasil clustering menunjukkan bahwa pemisahan data ke dalam dua klaster (k = 2) memberikan struktur yang cukup bermakna, dengan nilai Silhouette Score sebesar 0,17. Sistem ini berhasil diimplementasikan dalam bentuk aplikasi web berbasis Flask dan mampu memberikan rekomendasi mobil yang relevan berdasarkan kemiripan fitur dalam klaster yang sesuai.
====================================================================================================================================
The used car market in Indonesia has experienced significant growth along with technological advancements and the increasing use of e-commerce platforms such as OLX.co.id. The abundance of used car options with varying attributes, such as price, year of production, and fuel typem often makes it difficult for consumers to find a vehicle that matches their preferences. This study developed a used car recommendation system based on Content-Based Filtering (CBF) using the K-Nearest Neighbors (KNN) approach, which works by identifying similarities between user input and available car data. Prior to the recommendation process, the dataset was clustered using the K-Means algorithm to narrow the search space based on the similarity of numerical and categorical features. The clustering process showed that dividing the data into two clusters (k = 2) produced a reasonably meaningful structure, with a Silhouette Score of 0.17. The system was successfully implemented as a web application using the Flask framework and was able to generate relevant car recommendations based on feature similarity within the corresponding cluster.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Sistem Rekomendasi, Content-Based Filtering, K-Means Clustering, K-Nearest Neighbors, Mobil Bekas, OLX.co.id, Industry, Technological capabilities.Recommendation System, Content-Based Filtering, K-Means Clustering, K-Nearest Neighbors, Used Cars, OLX.co.id, Industry, Technological capabilities.
Subjects:	Q Science > QA Mathematics > QA76.9.I58 Recommender systems (Information filtering)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Farhan Himawan
Date Deposited:	28 Jul 2025 04:54
Last Modified:	28 Jul 2025 04:54
URI:	http://repository.its.ac.id/id/eprint/122598

Actions (login required)

View Item