Pengelompokan Produk Fast Moving Consumer Goods menggunakan Pendekatan Unsupervised Learning untuk Perbandingan Harga

Sri, Suci Indasari (2023) Pengelompokan Produk Fast Moving Consumer Goods menggunakan Pendekatan Unsupervised Learning untuk Perbandingan Harga. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 6026211007-Master_Thesis.pdf] Text
6026211007-Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (3MB) | Request a copy

Abstract

Perkembangan teknologi informasi serta perubahan gaya hidup masyarakat modern telah berdampak signifikan pada perkembangan layanan teknologi informasi, khususnya pada sektor pemasaran di lingkungan bisnis. Internet menawarkan peluang bisnis yang menjanjikan dengan melakukan penjualan produk kebutuhan sehari-hari khususnya produk Fast Moving Consumer Goods (FMCGs) melalui platform e-commerce denganberbagai penawaran produk dan harga yang terjangkau. Tiap e-commemrce memiliki karakteristik tersendiri sebagai upaya strategis dalam menarik perhatian konsumen, sehingga untuk memutuskan produk yang akan dibeli konsumen perlu membandingkan dengan e-commerce lainnya. Salah satu yang menjadi acuan keputusan pembelian konsumen yaitu harga produk. Sehingga, produk perlu disamakan agar harga dapat dibandingkan serta diurutkan. Untuk melakukan hal tersebut, produk harus dikelompokkan menurut kategori dan subkategori sebagai spesifikasinya. Sebanyak 2000 data produk berdasarkan lima kategori diolah untuk mengelompokan produk dengan Metode K-Means Clustering, data tersebut mengacu pada nilai groundthruth. Evaluasi data dilakukan menggunakan Confusion Matrix untuk membandingkan hasil clustering dengan nilai ground truth dan diperoleh 84% hasil clustering menyerupai. Sementara itu, penulis juga membangun suatu sistem perbandingan harga untuk memvisualisasikan proses pencarian produk multimarketplace dengan menerapkan information retrieval dalam pengimplementasiannya.
================================================================================================================================
The development of information technology and changes in the lifestyle of modern society have had a significant impact on the development of information technology services, especially in the marketing sector in the business environment. The Internet offers promising business opportunities by selling daily necessities, especially Fast Moving Consumer Goods (FMCGs) products through e-commerce platforms with various product offers and affordable prices. Each e-commerce has its own characteristics as a strategic effort to attract the attention of consumers, so to decide which products to buy consumers need to compare with other ecommerce. One of the references to consumer purchasing decisions is the price of the product. Thus, products need to be equalized so that prices can be compared and sorted. To do so, products must be grouped by category and subcategory as specifications. A total of 2000 product data based on five categories were processed to group products with the K-Means Clustering Method, the data refers to the ground truth value. Data evaluation was carried out using the Confusion Matrix to compare the clustering results with the ground truth value and obtained 84% of the clustering results resembling. Meanwhile, the author also built a price comparison system to visualize the multimarketplace product search process by applying information retrieval in its implementation.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Confusion Matrix, E-commerce, Fast Moving Consumer Goods, Information Retrieval, K-Means Clustering, Price Comparison, Confusion Matrix, Information Retrieval
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 59101-(S2) Master Thesis
Depositing User: Sri Suci Indasari
Date Deposited: 25 Jul 2023 02:12
Last Modified: 25 Jul 2023 02:12
URI: http://repository.its.ac.id/id/eprint/99470

Actions (login required)

View Item View Item