International Journal of Engineering and Computer Science Applications (IJECSA)

Analyze Threats in a Virtual Lab Network Using Live Forensic Methods on MetaRouter

2025-03-15T02:53:02+00:00

This research identified critical network anomalies in the MetaRouter virtual environment, focusing on IP activity related to routers, networks, and client devices. Suspicious interactions were observed between IP 192.168.1.100 (router) and IP 172.16.205.53 (client), including reused TCP port numbers and incomplete SYN sessions, indicating potential spoofing attempts. Invalid route information involving 192.168.1.100 highlights malicious modifications to the routing table, indicating an attempt to manipulate the routing information. Packet inconsistencies, such as “TCP Previous segment not captured” and “Spurious Retransmission,” revealed interference between the client and router, possibly caused by an external attacker exploiting network protocol vulnerabilities. The aim of this research is to analyze threats in virtual lab networks using live forensic methods on MetaRouter to detect anomalies, with a focus on Border Gateway Protocol (BGP) and TCP deviations in MetaRouter. This research method is a controlled prototype experimental setup in a virtual laboratory consisting of two routers and two client devices. This method simulates real-world network operations to identify malicious activities. Wireshark is used for real-time packet-level monitoring and analysis because it has powerful visualization and filtering capabilities, surpassing tools like tcpdump. This research integrates live forensic techniques to collect and analyze routing logs, packet data, and protocol behavior. The results of this research are the identification of suspicious behaviors, such as reused TCP port numbers, incomplete SYN sessions, and unauthorized route announcements, indicating potential spoofing and BGP hijacking attempts. Packet data irregularities, including “Out-Of-Order” messages and abrupt session terminations, are also detected, revealing disruptions in traffic flow caused by malicious activities. The results of this research are highlighting the effectiveness of the forensic framework in identifying and documenting network anomalies in virtual environments have significant implications for improving security in cloud-based and hybrid networks. This research provides a scalable and replicable methodology that can improve real-time anomaly detection and response, paving the way for future advances in network security.

Comparison of Lexicon-Based Methods and Bidirectional Encoder Representations for Transformers Models in Sentiment Analysis of Government Debt Market Movements

2025-03-15T02:53:02+00:00

The State Budget of Indonesia (APBN) is the main tool for implementing fiscal policies and serves as a budgeting guideline for development execution in Indonesia. One of the funding sources in budget financing is Debt Financing, which consists of Government Securities (SBN) issuance and Loans. Overall, SUN contributes IDR 5,824.34 trillion, highlighting its significant proportion in debt financing. Understanding public sentiment toward SUN is essential in developing effective government policies. This research conducts sentiment analysis on tweets from the social media X over the past 7.75 years to assess public perception and propose strategic recommendations. The aim of this research is to compare the BERT model and the Lexicon-Based method to determine which achieves the highest accuracy in sentiment analysis. The findings can help the government develop strategies for issuing SUN, especially in improving public involvement and investor trust. This research method is based on a deep learning pre-trained Bidirectional Encoder Representations from Transformers (BERT) model, specifically IndoBERT, with fine-tuning, and a Lexicon-Based approach utilizing the InSet lexicon. The results of this research are as follows: on the overall tweet dataset, the BERT model with optimal hyperparameters outperformed the Lexicon-Based method, achieving an accuracy of 70.28% compared to 55.77%. Similarly, on an annual basis, BERT exhibited higher accuracy than the Lexicon-Based method, except in 2021. Public sentiment on SUN in social media X is categorized as 49% positive, 30% neutral, and 21% negative. These findings indicate a generally favorable perception of SUN but also highlight areas for improvement in public communication. Overall, the BERT model demonstrates superior performance over the Lexicon-Based method. Considering the opportunities available, the government could leverage social media through Key Opinion Leaders and enhance transparency in explaining policies such as Tapera. This approach could maximize public participation in investing in SUN in Indonesia.

Optimization of Renewable Energy From Solar Panels For Environmental Monitoring Using Arduino

2025-03-15T02:53:02+00:00

Renewable energy is an alternative energy in the midst of the issue of fossil-based energy running out. One of these alternative energies is solar panels that generate electricity to achieve environmentally friendly energy. The problem researched in this study is the need to optimize or energy efficiency of solar panels. The type of research is development research with a prototype model. The purpose of this study is to optimize the efficiency of the solar panel tracking system equipped with artificial intelligence and sensors to monitor the environment based on the internet of things. This research method is an integration of fuzzy logic internet of things for tracking solar panels based on light sensor input. The use of solar panels using polycrystalline types, and the addition of temperature and humidity sensors is important to utilize. The result of this study is that the highest efficiency is obtained at 9.96%, meaning that the energy received by the system is 9.96% into useful power that is lost. In conclusion, solar panels are highly dependent on weather conditions and the focal point of solar energy capture. The implication of this study is that the improvement of environmentally friendly energy efficiency, although not too high, can be further developed. Recommendations for further research suggest that data collection be carried out in sunny weather, and improved using solar panels with a larger capacity and energy storage from batteries. The use of machine learning algorithms can be an alternative to study large amounts of solar panel data.

Analysist of User Satisfaction of the High School Student Admissions Website using the User Experience Questionnaire Method

2025-03-15T02:53:03+00:00

In the era of digitalization of public services, web-based student registration systems have become an important instrument in the education sector. The Website for New Student Admission (PPDB) of Senior High Schools (SMA) and Vocational High Schools (SMK) in Central Java Province has been implemented as a single platform for new student registration, but the main problem identified is the lack of a comprehensive evaluation of the level of user satisfaction with the quality of interaction experience with this platform, especially after the emergence of several complaints on the official PPDB social media regarding the system flow, services, and website appearance. The purpose of this study is to measure and analyze the level of user satisfaction with the PPDB website of SMA/SMK in Central Java Province using the User Experience Questionnaire (UEQ) approach which covers six aspects of user experience. This research method is descriptive quantitative with a survey approach using a standardized UEQ instrument consisting of 26 question items, involving 30 respondents of class X students of SMA Negeri 1 Karanganyar Demak selected using a 10% sampling technique from the population. The results of this study are indicate that the efficiency criteria obtained the highest score of 1.125, while the novelty criteria received the lowest score of 0.792, with the benchmark comparison diagram indicating a position below average (poor) in the criteria of attractiveness (1.061), clarity (1.092), accuracy (0.983), and stimulation (0.992), while in the criteria of efficiency and novelty, they are in a position above average (quite good). The implication of these findings underlines the need for further development in the aspects of visual appeal, clarity of information, accuracy of functions, and interaction stimulation to improve the overall quality of the user experience of the PPDB website.

Prediction of Student Major Selection at High School Using a Machine Learning Approach

2025-03-26T03:09:21+00:00

The primary objective of this research was to develop and evaluate a machine learning prediction system that matches Senior High School (SMA) Nusa Putra Kota Tangerang students with their potential school majors based on their academic interests and performance levels. This research method employs machine learning algorithms, including Random Forest, Support Vector Machine (SVM), logistic regression, K-Nearest Neighbor (K-NN), and Naive Bayes, using academic records, interest tests, and questionnaires for data collection. The data was processed and analyzed to train and test the algorithm. The findings of this study indicate that the Random Forest algorithm achieved the best performance among the models, with an accuracy of 85%, a precision of 82%, a recall of 88%, and an AUC score of 0.92. The factors that affected the prediction of major selection were Grade XII Mathematics scores and Science Interest Test results. The research implications suggest that Random Forest technology within Machine Learning (ML) enhances major selection accuracy while promoting fairness, providing superior educational choices and increased student satisfaction. Future studies should investigate additional factors that influence this phenomenon.