Implementasi algoritma jaccard similarity dan haversine dalam perancangan sistem marketplace supermarket

Tania, Calvine (2022) Implementasi algoritma jaccard similarity dan haversine dalam perancangan sistem marketplace supermarket. Bachelor thesis, Universitas Pelita Harapan.

Preview	Text (Title) Title.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (36kB) | Preview
	Text (Abstract) Abstract.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (76kB)
	Text (ToC) ToC.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (102kB)
	Text (Chapter1) Chapter1.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (153kB)
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (358kB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (73kB)
	Text (Bibliography) Bibliography.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (71kB)
	Text Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (959kB)

Abstract

People generally carry out the monthly shopping process on weekends, this causes supermarkets to become very crowded and causes difficulties and spends quite a long time in the shopping process. Although there are several systems that provide a monthly shopping process, these websites still have shortcomings such as not being able to provide a wider selection of goods and the process of searching for goods is still conventional and there is no application of product recommendations in it. Because of these problems, it is necessary to build an E- Commerce system that applies the B2B (Business to Business) business model or often known as the marketplace. The marketplace business model will be applied to supermarkets so that it can solve the problems in this research. In this research, two recommendation algorithms will be applied, namely the Jaccard Similarity algorithm to recommend favorite products and Haversine to recommend the nearest supermarket. This study chose the Jaccard Similarity algorithm because this algorithm is quite simple and accurate when applied to the recommendation case. Meanwhile, the Haversine algorithm was chosen because its formula is simple but suitable for calculating distances. The results show that the Jaccard Index and Haversine methods are proven to be accurate in providing product and supermarket recommendations that are closest to the user's position and the supermarket marketplace application that was built can solve the problems in this study./ Masyarakat umumnya melakukan proses belanja bulanan pada akhir pekan, hal ini menyebabkan supermarket menjadi sangat ramai dan menyebabkan kesulitan serta menghabiskan waktu yang cukup lama dalam melakukan proses belanja. Meskipun terdapat beberapa sistem yang menyediakan proses pembelanjaan bulanan, namun website-website tersebut masih memliki kekurangan seperti belum bisa memberikan pilihan barang yang lebih banyak serta proses pencarian barang masih berbasis konvensional dan belum adanya penerapan rekomendasi produk di dalamnya. Oleh karena permasalahan tersebut, maka perlu dibangun sebuah sistem E-Commerce yang menerapkan model bisnis B2B (Business to Business) atau sering dikenal dengan marketplace. Model bisnis marketplace akan diterapkan pada supermarket sehingga dapat menyelesaikan permasalahan pada penelitian ini. Pada penelian ini, akan diterapkan dua buah algoritma rekomendasi yaitu algoritma Jaccard Similarity untuk merekomendasikan produk favorit dan Haversine untuk merekomendasikan supermarket terdekat. Penelitian ini memilih algoritma Jaccard Similarity kareana algoritma ini cukup sederhana dan akurat jika diterapkan pada kasus rekomendasi. Sedangkan, algoritma Haversine dipilih dikarenakan formulanya yang sederhana namun cocok dalam penghitungan jarak. Hasil penelitian menunjukkan bahwa metode Jaccard Index dan Haversine terbukti akurat dalam memberikan rekomendasi produk dan supermarket yang paling dekat dari posisi pengguna serta aplikasi marketplace supermarket yang dibangun dapat menyelesaikan permasalahan pada penelitian ini.

Actions (login required)

View Item