ComTech: Computer, Mathematics and Engineering Applications

Comparison of the Symmetric and Asymmetric Generalized Autoregressive Conditional Heteroscedasticity (GARCH) Models in Forecasting the 2018-2023 Jakarta Composite Index

Yenni Angraini — 2024-05-21

The Autoregressive Integrated Moving Average with Exogenous Variables (ARIMAX) method assumes a homogeneous residual variance, but data with high volatility can cause violations of this assumption. Hence, it is interesting to compare the forecasting accuracy of symmetric and asymmetric Autoregressive Conditional Heteroskedasticity (ARCH) models in various data conditions. The research aimed to compare the accuracy of the symmetric ARCH/ Generalized Autoregressive Conditional Heteroscedasticity (GARCH) and asymmetric TGARCH models in forecasting weekly Jakarta Composite Index (JCI) data on January 1^st, 2018, to April 24^th, 2023, by involving the influence of COVID-19 as a covariate variable and applying several validation scenario models to training and testing data. Based on the best-selected model, forecasting was carried out from May 1^st, 2023, to July 3^rd, 2023. The data used were weekly JCI opening data from January 1^st, 2018, to April 24^th, 2023, with the COVID-19 period as a covariate variable. The analysis results show that symmetric and asymmetric methods can handle violations of the heteroscedasticity assumption in the ARIMAX model. The best model produced based on four data validation scenarios is the asymmetric ARIMAX(3,1,3)-TGARCH(1,2) model with an average MAPE value of 3.158%. In this model, the COVID-19 variable significantly influences the JCI movement. Forecasting is done with forecasting results that are stable with confidence intervals that widen in each period.

Enhancing Consumer-to-Consumer (C2C) E-Commerce through Blockchain: A Model-Driven Approach

Aditiya Hermawan — 2024-05-21

The rapid progress of Information and Communication Technology (ICT), especially the Internet, has changed lifestyles in profound ways, including sharing ideas, virtual interactions, digital entertainment, and online transactions. It has resulted in businesses globally turning to electronic commerce (e-commerce) to market products. E-commerce has revolutionized operations with features such as centralized storage and detailed product information in the marketing process. However, the inefficiency and lack of transparency in these centralized systems lead to high costs and limited user control, posing a significant challenge. Challenges include commission fees from E-Commerce providers, which hinder business growth. The research aimed to propose a more efficient and transparent model for Consumer-to-Consumer (C2C) e-commerce using blockchain technology. The C2C model enhanced user transactions and minimized third-party dependence, but centralization increased costs and limited seller access. Thus, a decentralized blockchain approach was proposed for greater transparency in e-commerce. The research innovatively applied blockchain to C2C e-commerce, enhancing market efficiency and transparency. The research method applied was a combination of Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis and practical application. The result shows that the approach succeeds in reducing high costs, transparency of data storage, and dependence on third parties. Blockchain reduces third-party involvement and promotes a fairer business environment because it uses a tamper-proof database for transparency, security, and efficiency in the ever-growing e-commerce ecosystem. Blockchain ensures automated transactions, real-time data tracking, and data security.

Performance of Fuzzy C-Means (FCM) and Fuzzy Subtractive Clustering (FSC) on Medical Data Imputation

Sri Kusumadewi — 2024-05-22

Missing values or incomplete data are frequently encountered in medical records. These issues will be a serious problem if the data must be provided completely for analysis. The research aimed to prove the performance of the Fuzzy Subtractive Clustering (FSC) and Fuzzy C-Means (FCM) methods for solving imputation problems. Both methods were implemented using medical data. It had been conducted using K-Means as a crisp clustering approach for imputation. In the research, fuzzy clustering—a distinct methodology—was applied. The primary research contribution was the suggested fuzzy logic imputation method, which took uncertainty under consideration. The data sample consisted of patients who were at least 40 years old and had a history of hypertension, diabetes, heart disease, stroke, or chronic kidney disease. The test was carried out by taking random portions of data from the entire medical record. The randomization technique used a probability of 10%–50%. The results of the ANOVA test show that the p-value is greater than ∝(=0.05). It means that the imputed value does not differ from the original value, whether implemented in the FSC or FCM method. The algorithm’s performance is evaluated using the Pearson correlation coefficient. According to the t-test results, the FCM method has a higher correlation coefficient than the FSC method. It implies that FCM is superior to FSC.

Multiple Classifier System for Handling Imbalanced and Overlapping Datasets on Multiclass Classification

Dessy Siahaan — 2024-05-27

The performance of classification models suffer when the dataset contains imbalanced and overlapping data. These two conditions are already challenging separately and even more complex if they occur together. In the research, an ensemble method called a Multiple Classifier System was proposed to address these issues by combining K-Nearest Neighbour and Logistic Regression. The Synthetic Minority Oversampling Technique (SMOTE) method was also applied to balance the dataset. The One Versus One (OVO) decomposition technique helped the multiclass classification process. A simulation with 18 scenarios proves that the MCS-SMOTE model can handle these problems by providing good performance. The model’s performance is also tested using empirical data on Poverty in West Java in 2021. Empirical data also show that the proposed method performs well, with an accuracy rate of 80.09%, an F1 score of 0.782, and a G-Mean of 0.242. The areas with the highest poverty rates are Bogor, Bekasi City, Bandung City, Bekasi Regency, and Depok City, located near DKI Jakarta, the capital city. Based on existing predictor variables, poor households in West Java are more likely to occur when they do not have access to credit, the number of household members is more than three, multiple families live in one building, and the head of the household has not graduated from elementary school.

Water Quality Monitoring System Based on the Internet of Things (IoT) for Vannamei Shrimp Farming

Rosliana Eso — 2024-06-05

As Internet of Things (IoT) technology develops, water quality monitoring systems for Vannamei shrimp farms have become more inventive and straightforward. The prototype IoT system monitors and controls the pool using sensors that can measure water quality parameters, such as temperature, pH, and salinity. The research aimed to design an automated water quality monitoring system for Vannamei shrimp aquaculture. The research used the E-4052C sensor, DS18B20 sensor, and DFRobot V1.0 sensor as data transmitting hardware (transmitter) and the receiving hardware microcontroller NodeMCU ESP32 as data processing, management, and control system tools. Then, the system used a Wi-Fi network to transfer data from the microcontroller to the Message Queue Telemetry Transport (MQTT) server as a data cloud. Several software programs, including Telegram, Node-Red, and ThingSpeak, help Android devices display real-time data. Test results for the accuracy of the sensor’s reading on water pH are 99.71, with an error rate of 0.29%. Meanwhile, the accuracy of the temperature sensor is 98.03 with an error rate of 1.7%. On the other hand, the accuracy of the salinity sensor is 99.49, with an error rate of 0.41%. The results indicate that all sensors have excellent performance. The real-time monitoring display and Android Telegram notification functions are good, and the automatic water quality monitoring tool is successfully operating in the Vannamei shrimp pool in Pamandati, South Konawe District, Southeast Sulawesi ssProvince, Indonesia.

Psychological Stress Detection Using Transformer-Based Models

Derwin Suhartono — 2024-06-22

Stress is a significant mental health problem that results in a lack of concentration. It has been more widely identified through social media since people who are under stress usually post about their physical pain and tiredness. However, stress assessment through social media by professionals can be expensive and time-consuming. The research aimed to produce a stress detection system trained using a Twitter dataset to predict stress using the user’s input sentence. The experiments that were done in the research used transformer-based models such as Bidirectional Encoder Representations from Transformers (BERT) and Robustly Optimized BERT (RoBERTa). The research involved data pre-processing, model training, and model evaluation to ensure high-quality train data. Since the data were imbalanced, data trimming was performed in pre-processing to select data randomly until the balance matched. This process ensured the model’s effectiveness in the training and evaluation stages. The features used in these experiments were features from each pre-trained model. In evaluating the model, accuracy, loss, and F1 score were used as metrics. In the result, for BERT, accuracy reaches 0.848 with an F1 score of 0.847. Meanwhile, RoBERTa has an accuracy of 0.837 and 0.834. The results prove that BERT and RoBERTa can be used to classify stress with accuracy and an F1 score above 0.8. The experiment result shows that the BERT deep learning model can detect stress using the Twitter datasets.

Editorial Page and Table of Content

Evaristus Didik Madyatmadja — 2024-05-21