Fahimeh Ramezani

Dr. Fahimeh Ramezani

1. School of Computer Science, Faculty of Engineering and IT, University of Technology Sydney (UTS), Sydney, Ultimo, NSW 2007, Austalia
+1
orcid

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 June 2021 in Remote Sensing
Verified
Journal Article
Open access
PDF file
Mohsen NaderpourMohsen Naderpour
Mohsen Naderpour
HMHossein Mojaddadi
Hossein Mojaddadi
Fahimeh RamezaniFahimeh Ramezani
Fahimeh Ramezani
Forest fire is one of the foremost environmental disasters that threatens the Australian community. Recognition of the occurrence patterns of fires and the identification of fire risk is beneficial to mitigate probable fire threats. Machine learning techniques are recognized as well-known approaches to solving non-linearity problems such as forest fire risk. However, assessing such environmental multivariate disasters has always been challenging as modelling may be biased from multiple uncertainty sources such as the quality and quantity of input parameters, training processes, and a default setup for hyper-parameters. In this study, we propose a spatial framework to quantify the forest fire risk in the Northern Beaches area of Sydney. Thirty-six significant key indicators contributing to forest fire risk were selected and spatially mapped from different contexts such as topography, morphology, climate, human-induced, social, and physical perspectives as input to our model. Optimized deep neural networks were developed to maximize the capability of the multilayer perceptron for forest fire susceptibility assessment. The results show high precision of developed model against accuracy assessment metrics of ROC = 95.1%, PRC = 93.8%, and k coefficient = 94.3%. The proposed framework follows a stepwise procedure to run multiple scenarios to calculate the probability of forest risk with new input contributing parameters. This model improves adaptability and decision-making as it can be adapted to different regions of Australia with a minor localization adoption requirement of the weighting procedure.

9 March 2023 in Annals of Operations Research
Verified
Journal Article
Closed access
PDF file
MAMajid Azadi
Majid Azadi
Mehdi TolooMehdi Toloo
Mehdi Toloo
Fahimeh RamezaniFahimeh Ramezani
Fahimeh Ramezani
RSReza Farzipoor Saen
Reza Farzipoor Saen
FHFarookh Khadeer Hussain
Farookh Khadeer Hussain
HFHajar Farnoudkia
Hajar Farnoudkia

9 July 2019 in IEEE Transactions on Cloud Computing
Verified
Journal Article
Closed access
MAMajid Azadi
Majid Azadi
AEAli Emrouznejad
Ali Emrouznejad
Fahimeh RamezaniFahimeh Ramezani
Fahimeh Ramezani
FHFarookh Khadeer Hussain
Farookh Khadeer Hussain
An increasing number of organizations and businesses around the world use cloud computing services to improve their performance in the competitive marketplace. However, one of the biggest challenges in using cloud computing services is performance measurement and the selection of the best cloud service providers (CSPs) based on quality of service (QoS) requirements (Duan, 2017). To address this shortcoming in this article we propose a network data envelopment analysis (DEA) method in measuring the efficiency of CSPs. When network dimensions are taken into consideration, a more comprehensive analysis is enabled where divisional efficiency is reflected in overall efficiency estimates. This helps managers and decision makers in organizations to make accurate decisions in selecting cloud services. In the current study, variable returns to scale (VRS), the non-oriented network slacks-based measure (SBM) model and input-oriented and output-oriented SBM models are applied to measure the performance of 18 CSPs. The obtained results show the superiority of the network DEA model and they also demonstrate that the proposed model can evaluate and rank CSPs much better than compared to traditional DEA models.

6 May 2024 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
MSMohammad Siami
Mohammad Siami
Mohsen NaderpourMohsen Naderpour
Mohsen Naderpour
Fahimeh RamezaniFahimeh Ramezani
Fahimeh Ramezani
JLJie Lu
Jie Lu
Decision support systems (DSSs) are computer-based systems that support managers in making operational, tactical, and strategic decisions. DSSs have been built to assist with a wide range of applications; however, in this paper, we are primarily concerned with risk assessment, which helps decision-makers to evaluate the risk of events. In recent years, advancements in big data processing, artificial intelligence, and machine learning have provided new opportunities for businesses to use these technologies for risk assessment. Yet, using these techniques with massive unlabeled data in an uncertain situation is challenging. This paper presents an autonomous fuzzy decision support system (AFDSS) for risk assessment that uses advanced artificial intelligence, unsupervised learning, and fuzzy logic. The model learns from big data characterized by uncertainty and a lack of labels for maximum utility. In an innovative approach, fuzzy clustering first extracts events from big data. Then, the risks associated with those events are assessed via a fuzzy inference system. New events are subsequently predicted based on their similarity to previously evaluated events. Evaluations of AFDSS with a real-world insurance dataset containing 500,000 journeys by 2500 drivers show that the proposed model can consistently assess risk in the big data environment. These results were drawn from a sensitivity analysis where all input parameters were changed using optimistic, pessimistic, and neural strategies. Performance was good across all three categories.

1 July 2020
Verified
Conference Paper
Open access
Mohsen NaderpourMohsen Naderpour
Mohsen Naderpour
HMHossein Mojaddadi
Hossein Mojaddadi
Fahimeh RamezaniFahimeh Ramezani
Fahimeh Ramezani
Wildfire is one of the most frequent natural hazards across the globe, one which has cast a malevolent shroud over many communities in recent years, causing significant risk to human lives, infrastructure, and property. Wildfires are hydrogeological events which are bound to escalate, especially due to climate change. They are different from other natural hazards as they are mainly triggered by human interventions rather than natural triggers. The wildfire risk management is a complex process with many uncertainties in the assessment, fire behavior and spread modelling, and decision making. To predict wildfires, sophisticated temporal geospatial methods are required. This paper develops a wildfire probability prediction method considering the capabilities of Bayesian networks and fuzzy logic that can handle uncertainties and update probabilities in response to the availability of new data. The model takes into account the data from a geographic information system (GIS) for a specific area at micro level to estimate the wildfire probability and is able to update the probability due to any planned or unplanned changes in the area. Therefore, the proposed method can feed to future macro and micro risk-based decision-making situations in wildfire prone areas. The method is evaluated through a sensitivity analysis and its performance is investigated through a case study in New South Wales (NSW), Australia.

2 January 2017 in The Computer Journal
Verified
Journal Article
Closed access
PDF file
Fahimeh RamezaniFahimeh Ramezani
Fahimeh Ramezani
JLJie Lu
Jie Lu
JTJavid Taheri
Javid Taheri
Albert ZomayaAlbert Zomaya
Albert Zomaya
Virtual machine (VM) live migration has been applied to system load balancing in cloud environments for the purpose of minimizing VM downtime and maximizing resource utilization. However, the migration process is both time- and cost-consuming as it requires the transfer of large size files or memory pages and consumes a huge amount of power and memory for the origin and destination physical machine (PM), especially for storage VM migration. This process also leads to VM downtime or slowdown. To deal with these shortcomings, we develop a Multi-objective Load Balancing (MO-LB) system that avoids VM migration and achieves system load balancing by transferring extra workload from a set of VMs allocated on an overloaded PM to other compatible VMs in the cluster with greater capacity. To reduce the time factor even more and optimize load balancing over a cloud cluster, MO-LB contains a CPU Usage Prediction (CUP) sub-system. The CUP not only predicts the performance of the VMs but also determines a set of appropriate VMs with the potential to execute the extra workload imposed on the VMs of an overloaded PM. We also design a Multi-Objective Task Scheduling optimization model using Particle Swarm Optimization to migrate the extra workload to the compatible VMs. The proposed method is evaluated using a VMware-vSphere-based private cloud in contrast to the VM migration technique applied by vMotion. The evaluation results show that the MO-LB system dramatically increases VM performance while reducing service response time, memory usage, job makespan, power consumption and the time taken for the load balancing process.