International Journal of Informatics and Communication Technology (IJ-ICT)
Not a member yet
494 research outputs found
Sort by
A hybrid machine learning approach for improved ponzi scheme detection using advanced feature engineering
Get PDFPonzi schemes deceive investors with promises of high returns, relying on funds from new investors to pay earlier ones, creating a misleading appearance of profitability. These schemes are inherently unsustainable, collapsing when new investments wane, leading to significant financial losses. Many researchers have focused on detecting such schemes, but challenges remain due to their evolving nature. This study proposes a novel hybrid machine-learning approach to enhance Ponzi scheme detection. Initially, we train an XGBoost classifier and extract its features. Meanwhile, we tokenize opcode sequences, train a gated recurrent unit (GRU) model on these sequences, and extract features from the GRU. By concatenating the features from the XGBoost classifier and the GRU, we train a final XGBoost model on this combined feature set. Our methodology, leveraging advanced feature engineering and hybrid modeling, achieves a detection accuracy of 96.57%. This approach demonstrates the efficacy of combining XGBoost and GRU models, along with sophisticated feature engineering, in identifying fraudulent activities in Ethereum smart contracts. The results highlight the potential of this hybrid model to offer more robust and accurate Ponzi scheme detection, addressing the limitations of previous methods
The 360° beach video: a supporting mindfulness intervention with virtual reality
Get PDFThis article describes optimizing virtual reality (VR) with a 360° beach video model used for mindfulness interventions. Using VR with 360° beach videos to support the presence of an immersive environment can effectively support mindfulness practices. Students are interested in the integration of technology in school counseling. VR helps in creating immersive environments such as forests, beaches, waterfalls, etc. so that students focus more on practicing mindfulness and attention in the current moment. This article focuses on optimizing 360° beach videos in the breathing mindfulness process so that it helps bring out real experiences. Obstacles to practicing mindfulness include lack of focus, mind wandering and not concentrating. through the use of 360° beach videos with VR can increase focus and be more effective in mindfulness practice
Shellcode classification analysis with binary classification-based machine learning
Get PDFThe internet enables people to connect through their devices. While it offers numerous benefits, it also has adverse effects. A prime example is malware, which can damage or even destroy a device or harm its users, highlighting the importance of cyber security. Various methods can be employed to prevent or detect malware, including machine learning techniques. The experiments are based on training and testing data from the UNSW_NB15 dataset. K-nearest neighbor (KNN), decision tree, and Naïve Bayes classifiers determine whether a record in the test data represents a Shellcode attack or a non-Shellcode attack. The KNN, decision tree, and Naïve Bayes classifiers reached accuracy rates of 96.26%, 97.19%, and 57.57%, respectively. This study's findings aim to offer valuable insights into the application of machine learning to detect or classify malware and other forms of cyberattacks
Quality of service optimization for 4G LTE upload and download throughput
Get PDFDemand for mobile data services and people’s dependence on 4G LTE networks continue to increase. However, the quality of service (QoS) of this network still requires improvement, especially regarding the effect of QoS on throughput at specific frequencies. The research gap lies in the lack of indepth analysis of the impact of QoS parameters on network performance at frequencies of 2,100 MHz and 2,300 MHz. This study evaluates the effect of QoS parameters, such as delay, jitter, and packet loss, on throughput in 4G LTE networks at both frequencies. The research methodology uses an experimental approach with throughput, delay, jitter, and packet loss measurements in various network conditions. The results showed that delay (17.2174 ms to 37.0322 ms), jitter, and packet loss significantly influence throughput, ranging from 624.5 Kbps to 1,322.4 Kbps. The 2,100 MHz frequency tends to show better performance than 2,300 MHz. This study concludes that optimizing QoS parameters, especially delay and jitter, can significantly improve 4G LTE network performance. These findings provide practical contributions for mobile operators in improving network quality and customer satisfaction and open opportunities for further research on other frequencies or newer network technologies
Scaling of Facebook architecture and technology stack with heavy workload: past, present and future
Get PDFLeading social media Facebook has improved its architecture to meet user needs. Facebook has improved its systems to handle millions of users with heavy workloads and large datasets using innovative architectural solutions and adaptive strategies. The study examines Facebook’s architectural and technological advances in heavy workload and big data. To understand how Facebook scaled with a growing user base and data volume, history and system architecture will be examined. It will also examine how cloud storage and high-performance computing optimize resource utilization and maintain performance during peak user activity. Facebook is managing big data and heavy workloads with new technologies like the hybrid communication model that uses PULL and PUSH strategies for real-time messaging. Facebook switched from HBase to MyRocks for message storage to improve performance as data grew. Architectural scaling and technology stack research must prioritize data storage innovations and optimized communication protocols to handle heavy workloads and big data. The messenger Sync protocol reduces network congestion and improves synchronous communication, reducing resource consumption and maintaining performance under high load. High-performance computing (HPC) and cloud storage should be studied together to support complex compute workflows. This convergence may improve large-scale application infrastructures and encourage interdisciplinary collaboration for scalable and resilient systems
The bootstrap procedure for selecting the number of principal components in PCA
Get PDFThe initial step in determining the number of principal components for both classification and regression involves evaluating how much each component contributes to the total variance in the data. Based on this analysis, a subset of components that explains the highest percentage of variance is typically selected. However, multiple valid combinations may exist, and the final choice is often made manually by the researcher. This study introduces a novel yet straightforward algorithm for the automatic selection of the number of principal components. By integrating ANOVA and bootstrapping with principal component analysis (PCA), the proposed method enables automatic component selection in classification tasks. The algorithm is evaluated using three publicly available datasets and applied with both decision tree and support vector machine (SVM) classifiers. Results indicate that this automated procedure not only eliminates researcher bias in selecting components but also improves classification accuracy. Unlike traditional methods, it selects a single optimal combination of principal components without manual intervention, offering a new and efficient approach to PCAbased model development
Enhancing logo security: VGG19, autoencoder, and sequential fusion for fake logo detection
Get PDFThis paper deals with a way of detecting fake logos through the integration of visual geometry group-19 (VGG19), an autoencoder, and a sequential model. The approach consists of applying the method to a variety of datasets that have gone through resizing and augmentation, using VGG19 for extracting features effectively and autoencoder for abstracting them in a subtle manner. The combination of these elements in a sequential model account for the improved performance levels as far as accuracy, precision, recall, and F1-score are concerned when compared to existing approaches. This article assesses the strengths and limitations of the method and its adapted comprehension of brand identity symbols. Comparative analysis of these competing approaches reveals the benefits resulting from such fusion. To sum up, this paper is not only a major contribution to the domain of counterfeit logo detection but also suggests prospects for enhancing brand security in the digital world
An IoT-based approach for microclimate surveillance in greenhouse environments
Get PDFAs the demand for efficient and cost-effective greenhouse microclimate surveillance has increased, we developed a microclimate surveillance system using microcontroller technology that automatically collects temperature and relative humidity data and transmits it to a cloud server for remote surveillance and data analysis. 1971 microclimate data points were acquired over a 20-day evaluation period, providing insights into greenhouse environmental conditions with a negative linear regression between air temperature and relative humidity within the greenhouse and an R-squared of 0.73. This illustrates the system’s ability to record and quantify environmental conditions data. Additionally, we derived a predictive model to manage microclimate conditions using the regression formula y = -6.12X + 238.33, where X represents the air temperature and y represents the relative humidity. All the results show that the acquired data can be used to make decisions to optimize plant growth. The prototype we developed can be an alternative option for small and medium-sized farms that need a greenhouse surveillance system to improve operational efficiency and reduce surveillance costs. The system can be further developed by implementing additional sensors to monitor light intensity, wind speed, or soil moisture and further data analysis models to optimize greenhouse management
Cloud application design for financial reporting in Indonesia’s small and medium enterprises
Get PDFSmall and medium enterprises (SMEs) in Indonesia are increasingly developing, but the application of information technology (IT) in small medium businesses is still lacking because for small medium business owners, doing their own bookkeeping without a system will maximize profits. However, this makes bookkeeping ineffective and inefficient because it requires manual data input and reconciliation. Utilizing a cloud-based accounting information system (CAIS) can integrate data, increase productivity, and minimize infrastructure costs because there is no need to provide costs for physical infrastructure. In this research, CAIS was designed to produce financial reports that focus on small medium businesses in Indonesia. The method used is a qualitative method by conducting observations through literature study for data collection and the rational unified process (RUP) which is limited to the elaboration stage to produce a prototype design. So, the result of this paper is a system design that can be used as a guide to continue with system development. This system aims to simplify transaction records so that they can be more efficient and effective in producing financial reports. The use of CAIS is also expected to increase profits and maximize the use of internet and technology in small medium businesses
Load forecasting of electrical parameters: an effective approach towards optimization of electric load
Get PDFThe increasing need for energy and the increasing cost of electricity have prompted the development of smart energy optimization systems that can help consumers reduce their electricity consumption and minimize costs. These systems are developed on the concept of a “smart grid” which is a digitalized and intelligent energy network that provides help in the efficient distribution of energy. Load forecasting plays a crucial role in the precise prediction of uncontrollable electrical load. Long-term load analysis predicts a load of more than one year and helps in the planning of power systems whereas short-term and medium-term load forecasting helps in the supply and distribution of load, maintenance of load system, ensuring safety, continuous electricity generation, and cost management. Machine learning (ML) focuses on the development of smart energy optimization systems by enabling intuitive decision-making and reciprocation to sudden variations in consumer energy demands. This study focuses on the consumption of consumer electricity and provides a solution regarding the optimized methods that will predict future consumption based on previous data and help in reducing costs and preserving renewable energy. This research promotes sustainable energy usage. The use of ML models enables intelligent decision-making and accurate predictions, making the system an effective tool for managing electricity consumption