This is a single portfolio entry
Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Aenean commodo ligula eget dolor. Aenean massa. Cum sociis natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Donec quam felis, ultricies nec, pellentesque eu, pretium quis, sem. Nulla consequat massa quis enim.
Donec pede justo, fringilla vel, aliquet nec, vulputate eget, arcu. In enim justo, rhoncus ut, imperdiet a, venenatis vitae, justo. Nullam dictum felis eu pede mollis pretium. Integer tincidunt. Cras dapibus. Vivamus elementum semper nisi. Aenean vulputate eleifend tellus. Aeneaellus.
Recent Entries
L131115121126
This paper selects an appropriate machine learning model to forecast real estate and apartment prices across several major provinces and cities in Vietnam. While we primarily deploy the Transformer, we also experimented with other machine learning algorithms, such as Linear Regression, Decision Tree, and Random Forest, to demonstrate the Transformer’s superior performance. We automatically collected raw data from Vietnam’s largest real estate website and then preprocessed, analysed, and evaluated it. The analysis revealed that building a prediction model on a single dataset yields fewer effective results than splitting the data into two distinct sets, one for landed properties and the other for apartments, and developing separate models for each product category. Building two independent prediction models allows effective use of each product type’s unique characteristics, improving the accuracy and efficiency of machine learning for real estate price prediction.
L131015121126
Intelligent, sustainable management systems are needed to address rapid urbanisation and rising energy consumption in commercial facilities. Traditional Building Management Systems (BMS) are reactive, non-predictive, and non-automated, and they lack AI support for decision-making. This paper introduces EcoMind, an Internet of Things (IoT)- based intelligent energy optimisation framework for smart commercial infrastructures, supported by artificial intelligence. It is implemented in a cloud-native system that incorporates IoT telemetry simulation, predictive analytics, Digital Twin monitoring, sustainability assessment, and a telemetry-based conversational AI module. Anomaly detection, predictive maintenance, occupancy-aware intelligence, and rule-based optimisation mechanisms provide real-time analysis of HVAC, lighting, elevator, escalator, parking, and environmental sensor data. The definitions used for this study are as follows: occupancy utilisation: ratio of actual occupancy over maximum occupancy; total energy consumption: total energy consumed by HVAC, lighting, elevator, escalator and parking systems; anomaly deviation: dif erence between the current sensor reading and the historical normal behaviour; optimisation efficiency: ratio of reduction in energy consumption after applying optimisation strategies to the total energy consumption. The major innovation is using the Gemini API’s large language model to integrate live infrastructure telemetry data into a pipeline, enabling real-time reasoning without relying on static datasets. Experimental results show a decrease in total energy consumption of 18–22%, improved response to anomaly detection and improved operational decisionmaking. The results confirm that Digital Twin intelligence and Telemetry AI is a powerful complement that enhances scalability, adaptability, and sustainability in smart commercial infrastructure systems.
K130915111026
Digital twin (DT) technology combined with artificial intelligence (AI) is becoming a key enabler for resilient and self-optimising electrical power networks that are increasingly digitalised. A digital twin is a continuously updated virtual representation of a physical grid asset or network that uses real-time sensor, phasor measurement unit (PMU), and Internet of Things (IoT) data to closely mirror its physical counterpart. Digital twins become predictive, prescriptive, and self-healing decision-support systems when machine learning (ML) and deep learning (DL) algorithms are integrated into this virtual-physical feedback loop. This review aims to bring together the recent literature (2020-2025) on the field of AIpowered digital twins under four thematic areas: (i) architectural foundations and enabling technologies, (ii) real-time monitoring and state estimation, (iii) predictive control and maintenance, and (iv) cybersecurity monitoring and threat mitigation. Tables summarise enabling technologies, predictive-control case studies, and case studies of cyber-threat countermeasures reported in the literature. The review also pinpoints open research challenges such as interoperability standards, federated and privacypreserving learning, explainability of AI-driven decisions on the grid, and quantum-assisted twin computation. It suggests a research roadmap for further work. The synthesis signals that while AI-powered digital twins can tangibly reduce unplanned downtime, improve forecasting precision, and boost cyberphysical attack resilience, standardisation, data governance, and trustworthy AI remain to be addressed to enable widespread adoption.
K130815111026
Modern times require smart solutions for crime detection to enhance public security levels because crime detection remains a fundamental challenge. Traditional surveillance relies on human monitoring but has weaknesses due to time consumption and human error. The automated crime detection system runs on machine learning and deep learning platforms using Convolutional Neural Networks and Recurrent Neural Networks, along with YOLO, Faster R-CNN, and the Azure Face API. The AI models perform two functions: detecting suspicious activities and identifying weapons and faces, and recognising patterns of criminal behaviour. The combination of local servers and cloud platforms powered by AI detects crimes precisely in real time through enhanced monitoring, improves accuracy, and prevents false alarms, while also enabling broader crime-prevention capabilities.
