Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Building of Informatics, Technology and Science

Prediksi Persediaan Bahan Baku Makanan Menerapkan Algoritma Apriori Data Mining Salmon, Salmon; Azahari, Azahari; Yusnita, Amelia
Building of Informatics, Technology and Science (BITS) Vol 4 No 3 (2022): December 2022
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v4i3.2563

Abstract

The company's operational activities are inseparable from the supply of raw materials that must be met every day to meet consumer demand. The restaurant uses raw materials, namely vegetables, raw meat which includes beef and chicken, yellow noodles and soun noodles, and the main seasoning. Sales of food at this restaurant quite a lot in a day. This will produce sales data that will continue to grow every day, but this data is useless if it is not processed again to get the knowledge contained in the data. The Apriori algorithm is a method for finding patterns of relationships between one or more items from a dataset. Thus the pile of data that has been collected can produce a sales pattern, from which the customer's buying interest in food can be identified. From the results of research using a data sample of 18 items with a minimum of 20% Support and 50% Confidence, it produces 5 interesting rules with the highest Support reaching 33.33% and the highest Confidence reaching 100%.
Clustering the Economic Conditions of Various Countries From 2010-2023 by Conducting A Comparative Analysis of the K-Means and K-Medoids Algorithms Yunita, Yunita; Ekawati, Hanifah; Yusnita, Amelia
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7330

Abstract

Understanding the similarities and differences in economic conditions across countries is crucial for various stakeholders. This research investigates the global economic landscape by clustering countries based on their economic indicators, including GDP, inflation rate, unemployment rate, and economic growth, spanning the period of 2010 to 2023. This timeframe encompasses significant global economic events, making it pertinent for analysis. The study employs and compares two prominent clustering algorithms: K-Means and K-Medoids, to identify groups of countries exhibiting similar economic patterns. Utilizing secondary data from Kaggle encompassing 19 countries, the research assesses the ability of each algorithm to delineate meaningful economic clusters. The K-Means algorithm, with a determined optimal number of four clusters, demonstrated a reasonably good cluster separation and moderate internal cohesion, evidenced by a Silhouette Coefficient of 0.58 and a Davies-Bouldin Index of 0.63. In contrast, the K-Medoids algorithm yielded a distinct clustering structure with a lower Silhouette Coefficient (0.26) and a higher Davies-Bouldin Index (1.16), suggesting less distinct cluster separation and potential sensitivity to data characteristics. This comparative analysis provides insights into the applicability and performance of K-Means and K-Medoids in discerning global economic structures, contributing to a deeper understanding of the world economic map and the utility of clustering techniques in economic data analysis.