PREDIKSI PENJUALAN TRENDING PAKAIAN DI TOKO FANTASTIK BERBASIS WEB MENGGUNAKAN METODE K-NEAREST NEIGHBOR
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Putra, Rizky Pradana Audy Setya
Putra, Rakhmadi Irfansyah
Yosrita, Efy
Putra, Rakhmadi Irfansyah
Yosrita, Efy
Subject
Teknik Informatika
Datestamp
2022-09-27 07:06:14
Abstract :
Clothing is a basic need other than food and shelter, which humans are definitely looking for. Consumption of this need for clothing will continue to increase in line with increasing population growth. As is the case in Nganjuk Regency, East Java. Toko Fantastik is a clothing store in Nganjuk Regency. It is not surprising that shop owners must be smart in reading market trends all the time so that there is no accumulation of clothing stocks in their warehouse. If the store management is not good, it will have an impact on profit and reduce revenue. Therefore, it is necessary to predict trends or market interest in web-based sales to make it easier for shop owners to analyze and manage clothing stocks at any time. This study uses the K-nearest Neighbor method, K-nearest Neighbor is a data mining algorithm that can be used to determine data predictions. How K-nearest neighbor works will compare test data and data sets, then take the distance to get the results from the nearest neighbor. The results of this study using the K-nearest Neighbor method with k = 3 to get clothing types of trousers, denim brands, categories of teenage buyers for women will be trending or sold a lot in 6 months for the next period. From the distance results then produce k (the closest neighbor) obtained, testing using the Confusion Matrix to find accuracy. And get optimal results of 75%.
Clothing is a basic need other than food and shelter, which humans are definitely looking for. Consumption of this need for clothing will continue to increase in line with increasing population growth. As is the case in Nganjuk Regency, East Java. Toko Fantastik is a clothing store in Nganjuk Regency. It is not surprising that shop owners must be smart in reading market trends all the time so that there is no accumulation of clothing stocks in their warehouse. If the store management is not good, it will have an impact on profit and reduce revenue. Therefore, it is necessary to predict trends or market interest in web-based sales to make it easier for shop owners to analyze and manage clothing stocks at any time. This study uses the K-nearest Neighbor method, K-nearest Neighbor is a data mining algorithm that can be used to determine data predictions. How K-nearest neighbor works will compare test data and data sets, then take the distance to get the results from the nearest neighbor. The results of this study using the K-nearest Neighbor method with k = 3 to get clothing types of trousers, denim brands, categories of teenage buyers for women will be trending or sold a lot in 6 months for the next period. From the distance results then produce k (the closest neighbor) obtained, testing using the Confusion Matrix to find accuracy. And get optimal results of 75%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
