Rekomendasi Barang Menggunakan Content Based Recommender System Berbasis Ethereum Blockchain

Putri, Shabrina Zhafarina (2023) Rekomendasi Barang Menggunakan Content Based Recommender System Berbasis Ethereum Blockchain. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 07211840000032-Undergraduate_Thesis.pdf] Text
07211840000032-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 July 2025.

Download (8MB) | Request a copy

Abstract

Pasar e-commerce Indonesia diperkirakan dapat menjadi kontributor pertumbuhan utama di Asia Pasifik. Berdasarkan analisis RedSeer, pasar e-commerce Indonesia diproyeksikan dapat meningkat menjadi US$137.5 miliar pada 2025(Pahlevi, 2022). Pada perkembangannya, e-commerce memerlukan suatu sistem rekomendasi. Sistem rekomendasi adalah aplikasi perangkat lunak yang menyediakan atau menyarankan item kepada pengguna yang dituju. Sistem ini menggunakan teknik penyaringan untuk memberikan rekomendasi. Motivasi tersebut muncul karena kebutuhan untuk mengintegrasikan fitur rekomendasi di e-commerce untuk mengurangi informasi yang berlebihan. Teknik berbasis konten diadopsi karena bersifat user independent. TF-IDF (Term Frequency Inverse Document Frequency) dan kesamaan kosinus digunakan untuk menentukan seberapa relevan atau miripnya item dengan item lainnya. Selain itu, e-commerce tidak lepas dari potensi ancaman yang disebabkan oleh peretas karena biasanya perusahaan e-commerce menyimpan datanya di lokasi terpusat (centralized) atau di cloud online. Data pengguna yang terbongkar secara langsung akan membuat kepercayaan pengguna terhadap situs tersebut menurun. Namun, masalah keamanan tersebut dapat dijawab melalui penggunaan teknologi blockchain.
=========================================================================================================================================
The Indonesian e-commerce market is expected to become a major growth contributor in Asia Pacific. Based on RedSeer’s analysis, Indonesia’s e-commerce market is projected to grow to US$137.5 billion in 2025(Pahlevi, 2022). In its development, e-commerce requires a recommendation system. Recommender systems are software applications that provide or suggest items to intended users. This system uses filtering techniques to provide recommendations. This motivation arises from the need to integrate recommendation features in e-commerce to reduce information overload. The content-based technique is adopted because it is user independent. TF-IDF (Term Frequency Inverse Document Frequency) and cosine similarity are used to determine how relevant or similar an item is to other items. In addition, e-commerce cannot be separated from potential threats caused by hackers because e-commerce companies usually store their data in a centralized location or in the online cloud. User data that is exposed directly will make users’ trust in the site decrease. However, these security issues can be answered through the use of blockchain technology.

Item Type: Thesis (Other)
Uncontrolled Keywords: Blockchain, Ecommerce, Sistem rekomendasi, Blockchain, E-commerce, Recommendation System.
Subjects: T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing
Divisions: Faculty of Electrical Technology > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User: Shabrina Zhafarina Putri
Date Deposited: 17 Feb 2023 12:27
Last Modified: 17 Feb 2023 12:27
URI: http://repository.its.ac.id/id/eprint/97555

Actions (login required)

View Item View Item