Komputika : Jurnal Sistem Komputer

Real-time Product Availability Information with Passive NFC Tag System for Offline Shops

Tati Erlina — Sat, 30 Dec 2023 00:00:00 +0700

Offering a seamless shopping experience is essential in today's cutthroat retail environment. This paper describes a system built to tell customers about product availability in offline stores, explicitly addressing the issues related to customer reluctance and the store's physical space restrictions. The system was designed, observed, and tested to evaluate its performance. According to the findings, the system can recognize various NFC tag types at reading distances of up to 3.5 cm, 1.5 cm, and 2 cm for a card, keychain, and sticker kinds, respectively. In an average of 3.39 seconds, the server and microcontroller can establish a connection to send and receive responses from the server. Additionally, the system has a 100% success record in displaying precise product stock data based on the chosen size and color. Furthermore, the system has a 100% success rate in telling registered from unregistered NFC tags. The internet network's speed also impacts updating database data, with quicker internet connections being processed first. In conclusion, the system's effectiveness demonstrates its potential to be used in retail settings to give customers real-time product availability tracking.

Comparison Analysis of K-Means and K-Medoids in Grouping Provinces Based on Indonesian Democracy Index 2021

Regita Dewanti Rudianto, Arie Wahyu Wijayanto — Sat, 16 Mar 2024 10:09:46 +0700

The clustering method is one method in data mining and is useful in grouping observations that do not have a target / class. One of the analyses that can be done from this clustering is the grouping of 34 provinces in Indonesia based on aspects in the 2021 Indonesian Democracy Index (IDI). The aspects of the IDI include the Freedom Aspect, Equality Aspect, and the Capacity Aspect of Democratic Institutions. Clustering analysis needs to be done to determine the grouping of IDI aspects and their characteristics. The clustering methods used in this study are K-Means and K-Medoids. For the selection of the optimal number of clusters used Dunn Index, Silhouette Index, Calinski-Harabasz Index and Davies-Bouldin Index. To obtain the best model, a comparison is made using the ratio between average within (Sw) and average between (Sb). The results obtained are that there are 5 clusters in the IDI grouping using the K-Medoids algorithm because the ratio of Sw/Sb is smaller than K-Means. With this grouping, it is hoped that the government and related parties can utilize the results of this analysis in formulating policies and maintaining political stability in Indonesia.

Multi-Aspect Sentiment Analysis of Film Review Using Bidirectional Encoder Representations from Transformers (BERT)

Nur Karimah — Sat, 16 Mar 2024 14:32:07 +0700

This research was conducted to apply the Bidirectional Encoder Representation from Transformer (BERT) method to multi-aspect sentiment analysis of film reviews. The review data was obtained using the scraping method. The dataset used consists of 1899 data to 3245 data having a positive sentiment, 4825 data with a neutral sentiment, and 1424 data with a negative sentiment. The proposed approach includes the aspects such as acting, plot, cast, animation, and music. The aspect with the most positive sentiment is music with a total of 631 data, the neutral sentiment is found in the animation aspect with a total of 1146, and the negative sentiment is found in the plot aspect with a total of 362. The dataset used went through cleaning data, including case folding and removing HTML tags, punctuation, numbers, and special characters. This research uses the BERTBASE-UNCASE model with four experiments using hyperparameters max_epoch 10, batch size 16, and learning rates of 1e-4, 5e-5, 3e-5, and 2e-5. The research results show that, from all experiments, the best accuracy value is achieved in the third experiment using a learning rate of 3e-5, which is 82,32%. Meanwhile, the best precision, recall, and f1-score values for the “animation” aspect are 86%, 85%, and 85%.

Pengenalan Huruf BISINDO Menggunakan Chain Code Contour dan Naive Bayes

Mulianty Cipta Irja, Dolly Indra, Lilis Nur Hayati — Fri, 22 Mar 2024 10:09:35 +0700

Digital image processing, also known as digital image manipulation, is a method used to process or manipulate digital images. Digital image processing can address various problem domains, one of which is the recognition of Indonesian Sign Language (BISINDO) letters used by the deaf and speech-impaired individuals for communication. The aim of our research is to develop a digital image-based application that can recognize BISINDO letters from A to Z with a high level of letter similarity accuracy. The BISINDO letter dataset consists of 260 images, divided into an 80% (208 images) training data set and a 20% (52 images) testing data set. The letter recognition process begins with pre-processing, including converting RGB images to grayscale, segmentation using thresholding, morphological opening, and Sobel edge detection. The shape feature extraction is then performed using Chain Code Contour. The values obtained from this feature extraction are used in the final stage, which is the recognition of BISINDO letter images using the Naive Bayes classification method. The research involves two testing scenarios: a database scenario and an out-of-database scenario, each with three dataset divisions: 80:20, 70:30, and 60:40. The results of the database scenario testing with an 80:20 dataset division achieved 100% accuracy, while the 70:30 division achieved 92.3% accuracy, and the 60:40 division achieved 88.4% accuracy. In the out-of-database scenario, the 80:20 dataset division achieved 80.7% accuracy, the 70:30 division achieved 73.07% accuracy, and the 60:40 division achieved 75.9% accuracy. Based on the conducted testing, the best accuracy was obtained with the 80:20 dataset division, achieving 100% accuracy in the database scenario and 80.7% accuracy in the out-of-database scenario. This indicates that the Chain Code Contour shape feature extraction method and Naive Bayes classification method are capable of recognizing BISINDO letters effectively.

Penerapan Metode Random Forest dalam Klasifikasi Huruf BISINDO dengan Menggunakan Ekstraksi Fitur Warna dan Bentuk

Mega Asfirawati Daris, Dolly Indra, Lilis Nur Hayati — Fri, 22 Mar 2024 10:20:49 +0700

Digital image processing is a field of study that focuses on how an image can be formed, processed, and analyzed to generate useful information for humans. In this research, the utilization of digital images is implemented to classify BISINDO (Indonesian Sign Language) letters from A to Z using the Random Forest classification method. The initial stage in the classification of BISINDO letter images involves pre-processing, which includes converting RGB images to grayscale and performing segmentation through three stages: thresholding, morphology, and edge detection using the Prewitt operator. Subsequently, features such as HSV color extraction and metric shape features, as well as eccentricity, are extracted. These extracted feature values are then utilized in the classification stage of BISINDO letter images from A to Z using the Random Forest method. In this study, three data comparison scenarios were employed for testing purposes. The first scenario involved an 80:20 data ratio, which achieved a testing accuracy of 94.2%. The second scenario with a 70:30 data ratio achieved a testing accuracy of 93.6%, while the third scenario with a 60:40 data ratio had a lower accuracy of only 77.9%. Based on the results of our testing, the system developed is capable of effectively classifying BISINDO letters from A to Z using color and shape feature extraction, along with the Random Forest classification method. The best results were obtained in the data comparison scenario of 80:20, achieving an accuracy rate of 94.2%.

Keywords – BISINDO, HSV, Metric, Eccentricity, Random Forest.

Pemodelan Clustering Ward, K-Means, Diana, dan PAM dengan PCA untuk Karakterisasi Kemiskinan Indonesia Tahun 2021

Kautsar Hilmi Izzuddin, Arie Wahyu Wijayanto — Mon, 01 Apr 2024 07:58:08 +0700

Poverty is a serious and quite complex problem. Poverty is influenced across sectors from various factors. Poverty grouping can be done for planning and evaluating poverty programs. Cluster analysis using the ward, k-means, diana, and PAM methods can be used to group provinces in Indonesia based on six poverty indicators, namely the percentage of poor people (P0), poverty depth index (P1), poverty severity index (P2), Open Unemployment Rate (TPT), Literacy Rate (AMH), and Average Years of Schooling (RLS). Based on the evaluation of the model, the best cluster model was obtained using the ward approach with Principal Component Analysis (PCA) analysis. PCA is proven to be able to maximize the performance of clustering models. The cluster ward model forms five optimal clusters with provinces with very low to very high poverty rates.

Investigasi Model Machine Learning Terbaik untuk Memprediksi Kemampuan Penghambatan Korosi oleh Senyawa Benzimidazole

Muhamad Akrom, Cornellius Adryan Putra Sumarjono, Gustina Alfa Trisnapradika — Mon, 01 Apr 2024 08:16:17 +0700

This research aims to investigate the corrosion inhibition performance of Benzimidazole compounds using a machine learning (ML) approach. The main challenge in developing ML is to obtain a model with high accuracy so that the prediction results are relevant and accurate to the actual properties of a material. In this research, we evaluate various linear and non-linear algorithms to obtain the best model. Based on the coefficient of determination (R2) and root mean square error (RMSE) metrics, it was found that the AdaBoost Regressor (ADA) model was the model with the best predictive performance in predicting the corrosion inhibition performance of benzimidazole compounds. This approach offers a new perspective on the ability of ML models to predict effective corrosion inhibitors.

Design and Implementation of Bluetooth Low Energy Based Access Control System

Wed, 24 Apr 2024 00:00:00 +0700

This paper discusses the design of a Bluetooth Low Energy (BLE) based access control system intended to make access control more practical to implement on areas with high personnel turnover rate by making access rules easy to set and making access keys relatively safe to distribute compared to existing key-based access control systems. The use of BLE technology allows the system to estimate key position within the system by utilizing the curve fitting method for distance estimation and the trilateral method for positioning. The proposed system consists of a server, an admin panel, electronic locks, access keys in the form of a wearable, and access keys in the form of an Android application. The system is found to be capable of implementing access control functionalities and capable of implementing indoor positioning based on sections.

Pengembangan Aplikasi Mobile dan Alat Pendeteksi Kebakaran Berbasis Sensor untuk Keamanan Elektronik

Wed, 24 Apr 2024 08:33:14 +0700

The use of electronic devices has become common in today's era, both in industries and in everyday life. However, the potential fire hazards posed by these devices, such as gas stoves and industrial equipment, need to be taken seriously. Despite the installation and maintenance measures taken to reduce risks, there are still other factors that can trigger fires. Therefore, it is important to develop a fire detection tool that can provide information about the room conditions so that appropriate preventive measures can be taken before or during a fire incident. By utilizing temperature and humidity sensors like DHT22, as well as smoke sensors like MQ2, a fire detection device can be designed. Through the use of these sensors and suitable programming, the system is capable of providing real-time room condition information and sending notifications when significant changes occur. The experimental results have shown that the system is responsive to changes in room conditions and capable of providing early warnings regarding potential fires.

Analisis Cluster Kondisi Keterampilan, Akses dan Fasilitas Teknologi Informasi dan Komunikasi di Indonesia

Rahma watin, Noverlina Putri Permatasari, Arie Wahyu Wijayanto, Waris Marsisno — Wed, 24 Apr 2024 10:26:18 +0700

In facing the digital transformation era, there are still imbalances in terms of skills, access, and information and communication technology facilities in Indonesia. It is necessary to group areas to identify areas that are still lagging, as evaluation material for equitable development. The clustering of regions is done by comparing the Partitioning and Hierarchical Clustering Methods. The Partitioning Clustering algorithm used is K-Means Clustering, with an optimum number of clusters of 4. The Hierarchical Clustering algorithm used is Agglomerative Ward, with a coefficient value of 0.864. Grouping using the Agglomerative Ward method produces an optimum number of clusters of 3. The Hierarchical Clustering method is better than the Partitioning method, with a Silhouette Value of 0.37.

Analisis Cluster Provinsi di Indonesia Berdasarkan Pertumbuhan Ekonomi Tahun 202

I Kadek Mira Merta Ningsih, Arie Wahyu Wijayanto — Wed, 24 Apr 2024 00:00:00 +0700

Economic development is a central agenda that aims to develop a country's economy in a sustainable manner. Indonesia's economy in 2022 grew by 5,31 percent, higher than the achievements in 2021. Therefore, considering that the economy is a very crucial sector, equitable distribution of economic growth is an important thing to pay attention to for the equal welfare of the Indonesian people. Researchers conducted an analysis related to the grouping of economic growth conditions of provinces in Indonesia in 2022 using the K-Means, K-Medoids, Hierarchical and Fuzzy C-Means Clustering. The data used are 9 variables of economic growth in 34 provinces in Indonesia in 2022. The final result was obtained by the Hierarchical Ward method with 2 cluster as the best method based on the results of internal validation and stability validation. In this method, cluster 1 is obtained totaling 28 provinces while cluster 2 totaling 6 provinces. The characteristics of cluster 1 are high economic growth seen from the variable value of factors forming high HDI but still have a high open unemployment rate, while the characteristics of cluster 2, namely low economic growth, are known from the value of the gini ratio and a high percentage of poor people.

Optimasi K-Nearest Neighbor Dengan Particle Swarm Optimization Untuk Klasifikasi

Roudlotul Jannah Alfirdausy, Izzatul Aliyyah, Aris Fanani — Wed, 24 Apr 2024 12:34:36 +0700

ABSTRACT – Immune Thrombocytopenic Purpura (ITP) is a hematological disease caused by autoimmune damage to platelets, causing a person to bruise easily or bleed excessively. ITP disease must be detected early because it can cause chronic or long-term disorders, so this study aims to classify ITP disease in order to avoid misdiagnosis of patients and can be treated and treated immediately. This classification uses the PSO-KNN combination method. The results obtained from the classification using the PSO-KNN combination method are an accuracy value of 91.8% with an increase of 4.9% from the KNN standard, a sensitivity value of 91.2% with an increase of 11.8% from the KNN standard, and a specificity value of 92.6% with a decrease of 3.7% from the KNN standard. % The training and testing time of PSO-KNN is also faster than standard KNN so that PSO is able to optimize and improve the classification results of KNN.

Implementation of Analytical Hierarchy Process in Decision Support System for Selection of Quality Rice

Rismayani Rismayani — Wed, 24 Apr 2024 13:00:50 +0700

This study aims to process a Decision support system in selecting quality rice that fits the criteria and decision alternatives that apply the Analytical Hierarchy Process (AHP) and also builds a web-based system to process the data. The Logistics Affairs Agency (BULOG) is a state-owned company engaged in food logistics. These agencies include logistics/warehousing, surveys and eradication of pests, supply of plastic sacks, transportation business, trade in food commodities and retail industry. The method used is the Analytical Hierarchy Process (AHP) which produces a hierarchical order or ranking of alternatives, this system is expected to assist in the selection of quality rice. The result of the research is to build a web-based system by implementing AHP in determining quality rice and based on accuracy testing, 100% results are obtained. Based on the functional testing of the system using a black box valid results were brought, then in logic testing using a white box found data that was free from errors.