International Journal of Engineering and Computer Science Applications (IJECSA)

Determining and Managing Stock of Goods Based on Purchasing Patterns Using the Frequent Pattern Growth Algorithm

Muhammad Aldi Zarkhasy, Christofer Satria — Mon, 13 Nov 2023 20:58:59 +0800

Stock inventory is one of the important work activities, because stock inventory is the main element in the field of commerce. Mistakes regarding stock inventory will result in fatal problems, especially when stock inventory management is still done manually or improper stock inventory planning can cause the amount of stock to pile up due to the small amount of demand from consumers, resulting in the stock being damaged, especially those contains elements of expiration because it is not sold. On the other hand, if the stock of goods is low while consumer demand is high, consumers will turn to other supermarkets to look for the goods they want. This can make supermarkets lose money because they cannot meet consumer needs. Therefore, a system is needed to look for product item combination patterns using association techniques in determining and managing stock. This research aims to look for product item combination patterns in previous period transaction data based on purchasing patterns in determining and managing inventory at supermarkets using the FP-Growth method. Where with Min. Support 30% and Min. Confidence 70% produces 12 rules then with Min. Support 45% and Min.Confidence 60% produces 6 rules. Based on the comparison and analysis results obtained using the FP-Growth method, in this study the Min limit was chosen. Support 30% and Min. Confidence 70% due to generating 12 association rules. This information can be used as a reference for supermarkets in making decisions in determining and managing stock of goods based on purchasing patterns.

Handling Imbalance Data using Hybrid Sampling SMOTE-ENN in Lung Cancer Classification

Sun, 04 Feb 2024 00:00:00 +0800

The classification problem is one instance of a problem that is typically handled or resolved using machine learning. When there is an imbalance in the classes within the data, machine learning models have a tendency to overclassify a greater number of classes. The model will have low accuracy in a few classes and high accuracy in many classes as a result of the issue. The majority of the data has the same number of classes, but if the difference is too great, it will differ. The issue of data imbalance is also evident in the data on lung cancer, where there are 283 positive classes and negative classes 38. Therefore, this research aims to use a hybrid sampling technique, combining Synthetic Minority Over-sampling Technique (SMOTE) with Edited Nearest Neighbors (ENN) and Random Forest, to balance the data of lung cancer patients who experience class imbalance. This research method involves the SMOTE-ENN preprocessing method to balance the data and the Random Forest method is used as a classification method to predict lung cancer by dividing training data and testing 10-fold cross validation. The results of this study show that using SMOTE-ENN with Random Forest has the best performance compared to SMOTE and without oversampling on all metrics used. The conclusion is using the SMOTE-ENN hybrid sampling technique with the Random Forest model significantly improves the model's ability to identify and classify data.

Design of Fuel Monitoring Application for Reservoir Tanks in Army Fuel Supply Point on Military Logistics Corps Based on Internet of Things

Wed, 21 Feb 2024 00:00:00 +0800

The Bekang Corps, a vital component of the Army responsible for logistics and transportation, plays a crucial role in maintaining the military's operational readiness and mobility. In the intricate landscape of military operations, the seamless integration of strategy, logistics, and tactics becomes imperative for success. Integrated logistical support, encompassing maintenance, supply, personnel, education, and training, and base facility support, serves as the backbone of effective military operations. However, the manual monitoring of underground fuel tanks at Storage and Supply Points (SPBT) presents challenges in terms of potential errors, time consumption, and significant efforts. The aim of this research is to address these issues by focusing on leveraging Internet of Things (IoT) technology to design and implement a monitoring application specifically tailored for the SPBT environment within the Army's Bekang Unit. This research method is aimed at providing a real-time solution for efficiently monitoring and managing fuel levels. By integrating the Float Level Switch sensor and NodeMCU ESP 8266 microcontroller, this research establishes a foundation in IoT. The Android application, developed using Android Studio, serves as the user interface, while Firebase functions as the real-time database, facilitating seamless communication and data exchange. The results of this research are the successful implementation of this IoT-based solution, which not only enhances the accuracy and responsiveness of fuel level monitoring but also contributes significantly to military operational efficiency. The anticipated significant contribution of the application includes the enhancement of military operational efficiency, the reduction of human error risks, and an increased sense of responsibility regarding fuel availability for operational needs.

Exploring Customer Purchasing Patterns: A Study Utilizing FP-Growth Algorithm on Supermarket Transaction Data

Hairani Hairani, Juvinal Ximenes Guterres — Fri, 01 Mar 2024 11:42:33 +0800

The need to analyze consumer purchasing patterns using association techniques also lies in the increasingly fierce competition in the retail market. Supermarkets face the challenge of understanding their customers' buying patterns. By utilizing association techniques, supermarkets can identify customer buying trends and quickly and appropriately adjust their strategies. Thus, analyzing consumer purchasing patterns using association techniques is no longer an option but an urgent need for supermarkets that want to survive and thrive in a changing market. Therefore, this study aimed to analyze purchasing patterns in supermarkets using the FP-Growth method to understand purchasing behavior and identify relevant patterns from transaction data. The method used in this research was the FP-Growth association method to create association rules from customer transaction data. The findings of this research were the use of the FP-Growth method in analyzing supermarket customer purchasing patterns, which obtained 10 association rules for 2 itemsets and 11 association rules for 3 itemsets based on a minimum Support value of 30% and a minimum Confidence of 70%. The association rules generated by the FP-Growth method on 2 itemsets and 3 itemsets simultaneously bring up items often purchased by customers with the same pattern, namely Cooking Oil, Eggs, Flour, and Candy. This research concludes that the association rules formed can be used as a benchmark by supermarkets in preparing stock items and making strategies to increase sales for more profit.

Clustering Biplot on Tourist Visits in Indonesia

Isma Muthahharah, Zakiyah Mar'ah — Fri, 01 Mar 2024 22:42:42 +0800

This research aims to find out whether or not many tourists visited Indonesia after Covid-19 by clustering. This will generate foreign exchange earnings and contribute directly to the country's economic growth. The analytical method used in this research is K-Medoids. K-Medoids is a partition clustering technique that groups a collection of n objects into k clusters by utilizing the objects in the collection of objects to represent a cluster called a medoid. The data in this research used secondary data related to foreign tourist visits to Indonesia from several publication sources in 2017-2021. The results of this research show that there were 3 clusters obtained: Cluster 1 shows the number of tourist visits visiting Indonesia in 2017, 2018 and early 2019 because the Covid-19 pandemic has not yet occurred, Cluster 2 shows that there were no tourist visits in 2020 due to the start of the Covid-19 pandemic, and Cluster 3 indicates low tourist arrivals in 2021 due to the Covid 19 pandemic which temporarily prohibited foreign tourists from visiting Indonesia.