Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Jurnal Info Sains : Informatika dan Sains "Journal of Data Science
Anzas Ibezato Zalukhu
Unknown Affiliation
Automotive Engineering Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Analysis of Product Demand Prediction Using Decision Tree on Sales Data of Ceria Toys Store Anzas Ibezato Zalukhu; Muhammad Iqbal
Journal Of Data Science Vol. 3 No. 01 (2025): Journal Of Data Science, March 2025
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58471/jds.v3i01.6458

Abstract

Ceria Toys faces challenges in efficiently managing the inventory of electric bicycles, as product demand is influenced by factors such as market trends, seasons, and changing consumer preferences. To address this challenge, this research employs data mining techniques with the decision tree algorithm to predict product demand and assist in inventory management. The evaluation results of the predictive model show varying performance across product categories. The precision for the "Hot" category is 58.36%, while for the "Less Popular" category, it is 64.18%. The recall for the "Hot" category reaches 83.71%, but the recall for the "Less Popular" category is only 32.82%. Although the model performs better in predicting hot products, there is still room to improve the detection of less popular products. To enhance effectiveness, Ceria Toys can balance the dataset or adjust the model. With this information, the store can better prepare stock for hot products and optimize the management of less popular products. These steps are expected to maximize sales, reduce excess stock, and improve overall customer satisfaction.
Implementation of Data Mining for Mapping Student Visiting Interest Based on Study Programs as an Effort to Optimize Library Services at STMIK Methodist Binjai Anzas Ibezato Zalukhu; Adil Priman Hati Hulu
Jurnal Info Sains : Informatika dan Sains Vol. 16 No. 01 (2026): Info sains, 2026
Publisher : SEAN Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study successfully implemented the K-Means Clustering algorithm to map students’ library visit interest patterns at the STMIK Methodist Binjai Library. The clustering results were empirically validated using a Silhouette Coefficient score of 0.8304, with the optimal number of clusters determined as k = 3 through the Elbow Method. The clustering process identified three strategic profiles: a Low-Interest Cluster consisting of 117 students with an average of 1.50 visits, a Moderate-Interest Cluster comprising 19 students with an average of 9.32 visits, and a High-Interest Cluster including 8 students with an average of 21.00 visits. The data analysis revealed significant disparities in visit interest across academic programs, where students from the Informatics Engineering program demonstrated higher levels of engagement compared to those from the Information Systems program, which was predominantly characterized by low visit frequency. These findings provide a scientific foundation for library management to formulate segmented service optimization policies, including retention programs for active users and personalized literacy stimulation strategies to enhance student engagement in academic programs with lower visit intensity
Co-Authors Adil Priman Hati Hulu Muhammad Iqbal