Journal of Information and Technology

Remote Control Home Security Monitoring

2025-04-14T19:31:45+00:00

This research paper details the development and implementation of a Remote-Control Home Security Monitoring system designed to directly notify homeowners and detect living creatures entering their property. The primary objective is to provide timely information about intrusions, including capturing photographic or video evidence of the living entity, and offering the capability to deter entry using a camera-integrated speaker when necessary. This system aims to enhance home security by allowing remote monitoring of potential intrusions, providing peace of mind to homeowners when they are away. The core functionality of the system involves detecting living being movement using motion sensors and security cameras, ensuring real-time home surveillance. Upon detection, the system instantly alerts the homeowner via SMS messages and automated voice calls. Furthermore, it captures and stores video recordings of the detected living being for evidence and future review. The Remote-Control Home Security Monitoring system comprises several interconnected components: a Passive Infrared (PIR) sensor for motion detection, an Arduino Uno microcontroller for processing sensor data and controlling other devices, a buzzer and LED for local alerts, a GSM module for remote communication via SMS and calls, a V360Pro camera for video recording and potential remote interaction, jumper wires for electrical connections, glue for securing wiring, a power adapter, a smartphone or tablet for user interface, a laptop for Arduino code development, and necessary power cables. Testing results indicated successful initialization of all system components (Arduino Uno, GSM module, V360Pro camera, and PIR sensor) and establishment of internet connectivity for the camera. The GSM module successfully registered on the network and sent SMS messages. The PIR motion sensor accurately detected motion events with 100% accuracy in controlled testing, triggering the local alerts and the remote notification system. The V360Pro camera's live stream was accessible via the mobile application. The Remote-Control Home Security Monitoring system effectively integrates various hardware and software components to provide a comprehensive home security solution. It offers real-time detection of living creatures, immediate alerts to homeowners via multiple channels, and visual evidence through video recording. This technology holds significant potential for enhancing residential security and providing homeowners with increased control and reassurance regarding the safety of their property. The paper also suggests future research should focus on differentiating between anomalous and normal detections to further refine the system's intelligence and effectiveness in identifying unauthorized individuals. This research contributes to the advancement of smart home security systems and offers valuable insights for those seeking to implement such technologies for safeguarding their homes.

Applicability of Machine Learning and Internet of Things-Based for Crop Selection

2025-04-14T19:41:50+00:00

Agriculture plays a big role in ensuring global food security and sustainability. To optimize crop yields, it is essential to understand the soil and weather conditions of a farm. In this study, we developed a Crop Selection System that leverages the IoT and ML, specifically the Random Forest algorithm, to assist farmers of Rwanda in choosing the most suitable crops for their piece of land. The system makes recommendations based on measurements of nitrogen, phosphorus, potassium, pH, humidity, rainfall, and temperature before cultivation begins. Real-time field data was gathered using IoT sensors deployed in farm areas to collect soil and environmental information. Our device utilized soil sensors to measure nitrogen, soil pH, humidity, phosphorus, temperature, and potassium. The NodeMCU microcontroller preprocessed the data and uploaded it to a cloud database. The data collection took place in Gicumbi District, and the collected samples were analyzed using our crop selection model. The model was developed using the Random Forest algorithm to evaluate soil compatibility and rank the crops based on their probability of successful growth. By training the model with a dataset of crops’ ecological requirements, we achieved an accuracy of 96%. We then tested the model with newly collected data from the field. Over a period of seven days, the model's predictions indicated that potatoes had the highest growing probability at 56%, followed by beans at 43%, carrots at 33%, tomatoes at 33%, and rice at 17%. This data-driven approach significantly enhances farmers' decision-making by enabling them to make informed choices about crop selection. This technology can boost agricultural productivity while reducing unnecessary costs, such as excessive fertilizer use, by ensuring crops are cultivated based on the available ecological conditions. The system helps farmers assess nutrient levels and implement corrective measures to restore soil fertility. This method advances precision agriculture and contributes to the overall modernization of farming practices.

A Comparative Analysis of Ensemble-Based Models for Predicting Cryptocurrency Price Movements

2025-04-14T19:19:21+00:00

This study, "A Comparative Analysis of Ensemble-Based Models for Predicting Cryptocurrency Price Movements," evaluates ensemble machine learning models bagging, boosting, and stacking to improve cryptocurrency price prediction accuracy. Using historical data, models like Random Forest, Gradient Boosting, and Stacking were tested, with Stacking emerging as the top performer (81.80% accuracy, 81.49% F1-score, 88.43% AUC-ROC), outperforming traditional methods like Naive Bayes and Decision Trees. The Boosting Combined model also showed strong results. The research highlights the effectiveness of ensemble techniques in handling cryptocurrency market volatility, offering valuable insights for traders and investors. It underscores the potential of advanced feature engineering and real-time testing to further enhance predictive accuracy, advancing financial decision-making and risk management in the cryptocurrency sector.

AI Powered Network Traffic Detection

2025-04-22T17:14:41+00:00

This study presents an AI-powered network traffic detection framework capable of recognizing anomalies and addressing cyber threats in real-time. Traditional detection systems struggle to keep pace with evolving threats, necessitating more adaptive and intelligent approaches. To this end, the research integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) models to enhance detection accuracy and operational efficiency. The framework is evaluated using benchmark datasets such as UNSW-NB15 and CICIDS2017, focusing on performance metrics including accuracy, precision, recall, and false positive rate. Experimental results show the proposed hybrid model achieves a detection accuracy of 92.08%, with precision and recall exceeding 92%, and a low average detection latency of 0.00142 seconds per sample. These findings confirm the model's effectiveness in detecting both known and novel threats, making it a scalable and reliable solution for modern cybersecurity challenges. The system offers real-time threat mitigation and valuable insights for network administrators, contributing to more proactive and robust security postures.