Jurnal Online Informatika
Not a member yet
276 research outputs found
Sort by
Variational Quantum Circuit-Based Quantum Machine Learning Approach for Predicting Corrosion Inhibition Efficiency of Expired Pharmaceuticals
Full text linkThis study examines the potential of quantum machine learning (QML) to predict the corrosion inhibition capacity of expired pharmaceutical compounds. The investigation employs a QSPR model, using features generated from density functional theory (DFT) calculations as input. At the same time, corrosion inhibition efficiency (CIE) values obtained from experimental data serve as the target output. The VQC model demonstrates varied performance across evaluation metrics, especially with encoding and ansatz design. The model achieves fine scores in evaluation metrics, with root mean square error (RMSE) of 6.15, mean absolute error (MAE) of 5.63, and mean absolute deviation (MAD) of 5.50. The research underscores the significance of larger datasets for enhancing predictive accuracy and points to QML\u27s potential in exploring anti-corrosion materials. Although there are some limitations, this study provides a foundational framework for using QML to predict anti-corrosive properties
Synergistic Disruption: Harnessing AI and Blockchain for Enhanced Privacy and Security in Federated Learning
Full text linkCombining blockchain technology with artificial intelligence (AI) offers revolutionary possibilities for developing strong solutions that capitalize on each technology\u27s own advantages. Blockchain technology makes self-executing agreements possible by enabling smart contracts, which reduce the need for middlemen and increase efficiency by precisely encoding contractual terms in code. By using AI oracles, these contracts can communicate with outside data sources and make well-informed decisions based on actual occurrences. Additionally, there is a lot of potential for improving machine learning and data interchange in terms of privacy, security, and transparency through the integration of blockchain with federated learning. In order to provide accountability and transparency, the blockchain\u27s immutable ledger can painstakingly record every transaction that takes place during the federated learning process, from data submissions to model modifications and remuneration. Participants in federated learning networks also develop trust because of blockchain\u27s transparency and resistance to tampering. Strong participant verification procedures are put in place to strengthen data integrity and model updates, which raises the system\u27s overall reliability. In the end, this chapter examines novel research avenues for combining blockchain technology with federated learning, providing practical methods and strategies to improve transaction security and privacy and opening the door to a new era of reliable and effective machine learning applications
Forecasting Shallot Prices in Indonesia Using News-Based Sentiment Indicators
Full text linkThe volatile price changes of shallots are a challenge in controlling their prices. The fluctuation in the price of shallots is always reported in the media because it affects people\u27s lives. The news is released online via the internet and has beneficial information so it can be utilized. This study aims to provide a comparative analysis of forecasting models for shallot prices in Indonesia, evaluating the impact of using the most effective sentiment indicators derived from four lexicon-based methods. Data were collected by scraping method on three news portals and one food price information source website during the period from 2020 to 2023. The correlation and causality analysis was conducted to determine the relationship between food prices and sentiment indicators that was obtained using four sentiment analysis methods. The selected sentiment indicators for each day were used as an additional variable in forecasting using ARIMA, SARIMA, and BSTS models. The results showed that the use of news sentiment could reduce RMSE, MAPE, and MAE in forecasting shallot food prices.
Comparative Analysis of IndoBERT and LSTM for Multi-Label Text Classification of Indonesian Motivation Letter
Full text linkThe evaluation of motivation letters is a crucial step in the student admission process for one of vocational institutions in Indonesia. However, the current manual assessment method is prone to subjectivity and inconsistency, making it less reliable for fair student selection. This research presents a comparative analysis of two deep learning models, IndoBERT and Long Short-Term Memory (LSTM), for multi-label text classification of motivation letters written in Indonesian. Using a dataset of 676 motivation letters labeled with nine predefined categories, we evaluate the models based on their classification performance. The results indicate that IndoBERT outperforms LSTM, achieving an F1-score of 81%, compared to 76% for LSTM. This research provides insights into the effectiveness of IndoBERT for multi-label classification tasks in the Indonesian language and serves as a benchmark for future research in automating motivation letter evaluations
Performance Evaluation of NAS Parallel and High-Performance Conjugate Gradient Benchmarks in Mahameru
Full text linkHigh-Performance Computing (HPC) plays a crucial role in accelerating scientific advancement across numerous fields of research and in effectively implementing various complex scientific applications. Mahameru is one of the largest national HPC systems in Indonesia and has been utilized by many sectors. However, it has not undergone proper benchmarking evaluation, which is vital for identifying issues related to hardware and software configurations and confirming system reliability. Therefore, this study aims to evaluate the performance, efficiency, and capabilities of Mahameru. We present a benchmarking system on Mahameru utilizing two benchmark suites: the NAS Parallel Benchmarks (NPB) and the high-performance conjugate gradient (HPCG) benchmark. Our results indicate that the NPB exhibits a lower speedup in Message Passing Interface (MPI) compared to OpenMP, which can be attributed to the communication overhead and the nature of the computational tasks. Additionally, the HPCG benchmark demonstrates that Mahameru performance can compete with the lower tiers of the Top 500 supercomputers. When operating at full capacity, Mahameru can achieve approximately 2.5% of its theoretical peak performance. While the system generally performs reliably with parallel algorithms, it may not fully leverage hyperthreading with certain algorithms. This benchmark result can serve as a basis for decision-making regarding potential upgrades or changes to a system
Forensic Analysis of Web Scraping Documents on Carding Forums and Shops using Latent Dirichlet Allocation: Profiling Forensic and NLP Approaches for Cybercrime Investigation
No full textThis research is based on the massive cybercrime activity in carding forums and carding shops. Based on the many victims and losses from these activities a cybercrime investigation action is needed by a digital forensic investigator. The purpose of this study is to develop a forensic carding investigation framework based on document analysis of web scraping results on carding forums and carding shops, which applies forensic profiling analysis methods and natural language processing based on the latent dirichlet allocation (LDA) algorithm. The tools used for web scraping in this study are WebHarvy Version 7.3.0.222. The tools used for data processing in this study are Microsoft Excel and Orange Data Mining. The conclusion of this study shows that the application of web scraping investigation techniques on carding forums and carding shops based on an carding investigation framework has been effective in collecting relevant data and analyzing the activities of cybercriminal appropriately. Overall, this study has succeeded in developing a more organized and data-driven approach to dealing with crimes in carding forums and carding shops, which can be a reference for further research and application in the field of digital forensic investigation
Reviewing the Blockchain’s Framework and its Role in Sustainable Industries
Full text linkBlockchain technology is often regarded as a highly advanced and pioneering breakthrough in modern times. Blockchain technology is a distributed ledger that uses encryption to prevent security breaches and securely stores data across many systems. This facilitates collaborative transactions by providing a solitary, dependable reference point, revealing the purported trust intermediaries. This study aims to investigate the core principles of blockchain technology and assess its potential to support sustainability across various sectors. It seeks to examine how blockchain technology enhances reliability, effectiveness, and transparency in industries such as supply chain management and the energy sector. This study addresses these concerns by assessing the valuable applications, advantages, and drawbacks of blockchain in promoting sustainable industrial practices. Bitcoin and other cryptocurrencies rely on hashing as the foundation of their blockchain technology. Blockchain is a digital ledger that documents and tracks financial transactions. Blockchain technology has become prevalent across several sectors, encompassing artificial intelligence, machine learning, and the Internet of Things. Therefore, once the blockchain is prepared for dissemination, the data cannot be modified by anyone. This implies that it is immutable. Hyperledger offers a neutral platform for facilitating collaborative operations among organisations that frequently engage in competitive activities. Hyperledger is specifically designed to provide explicit support for blockchains as a means of business agreements. Authorisation is a prerequisite for a framework, ensuring that only those with proper authorisation can join the organisation. The ability of the manager to impose limitations on user access to the blockchain enhances security measures. Moreover, instead of being universally accessible through online platforms, trades are maintained secretly, limiting access to only essential participants. Using distributed code bases and open-source record upgrades facilitates enhanced efficiency in corporate activities. The fast expansion of blockchain technology has led to its widespread adoption across several industries worldwide. Illustrations encompass various domains, including logistics, copyright, finance, medicine, and supply chain management. Furthermore, we offer an introductory overview of blockchain technology, encompassing topics such as different types of blockchains and their utilisation across many sectors
Enhanced Agricultural Decision-Making: Machine Learning Approaches for Crop Prediction and Analysis in India
Full text linkThis paper addresses the critical aspects of agriculture in the Indian economy and the challenges faced by this sector, including soil quality decline, unpredictable weather, and the need for efficient decision-making. It presents machine learning as a transformative approach for improved agricultural decision-making, enabling enhanced crop prediction and productivity. Machine learning (ML) algorithms are shown to effectively analyze vast datasets to generate predictive models that aid in crop selection optimization, disease outbreak prediction, and market fluctuation anticipation, thus leading to increased yields and profitability. Focusing on crop prediction, the paper discusses models leveraging historical data and advanced algorithms to forecast crop yields. Additionally, the application of machine learning in precision farming, such as optimizing fertilizer application, is explored. The paper uses a mixed-method approach on a dataset encompassing various crops and environmental parameters. In this paper the various techniques such as K-Nearest Neighbor (KNN), Support Vector Machines (SVM), Decision Tree (DT) and Random Forest (RF) algorithms have been employed to demonstrate the utility of ML in the agricultural fields. The KNN at the value of K=4 and SVM with polynomial kernel resulted the accuracy of 0.982 and 0.989 respectively. Whereas DT and RT gave the results in terms of accuracy of 0.987 and 0.970 respectively. Overall, it can be said that all these techniques used in the present work showed the better accuracy for agricultural sustainability
Random Forest-Based Classification of Greywater Filtration Media for Intelligent Biofiltration Systems
Full text linkThe increasing volume of domestic wastewater, particularly greywater, has raised the demand for intelligent and adaptive treatment systems to support efficient water reuse. This study aims to develop a classification model for filtration media types (physical, chemical, and biological) based on water quality data using the Random Forest algorithm. Initial labeling was conducted using the K-Means Clustering method on a publicly available dataset simulated as greywater, based on ten key water quality parameters relevant to irrigation and environmental standards. Model evaluation demonstrated excellent classification performance, with a macro F1-score reaching 0.97 and consistent results in both 5-fold and 10-fold cross-validation. These findings indicate that the proposed model can be integrated into an IoT-based biofiltration system as an automated classification logic to support adaptive, efficient, and reusable household wastewater treatment in the context of irrigation
Pyramid Quantum Neural Network Based Resource Allocation with IoT: A Deep Learning Method
Full text linkAs more smart devices are connected and collecting massive quantities of data, the Internet of Things is growing rapidly. Resource management is another crucial issue since IoT networks are very diverse and often built and rebuilt dynamically. This study introduces a new kind of deep learning model known as the Pyramid Quantum Neural Network (PY-QNN) to solve the problem of resource allocation in Internet of Things systems. PY-QNN builds on quantum computing to improve the accuracy, scalability, and computation performance of Deep Learning. Because of superposition and entanglement, which increase generalization and provide faster convergence, QNNs enhance learning capabilities. The pyramid structure also helps manage the hierarchy of IoT networks. In order to forecast efficient resource assignment and implement this as soon as feasible to lower latency and boost efficiency, PY-QNN uses simulated resource and network requirements. Experimental findings demonstrate that PY-QNN outperforms baseline common deep learning techniques by reducing resource waste and offering online solutions, especially in large and complex IoT networks