Browsing by Author "Haque, Aminul"

Now showing 1 - 20 of 20

A Comparative Economic Approach to Maximize Profit for Providers in Grid Computing
(International Business Information Management Association (IBIMA), 2010-06-24) Haque, Aminul; Alhashmi, Saadat; Parthiban, Rajendran
Grid computing is a technology that shares computational power (such as storage, CPU and memory) from distributed sites across the globe. Due to the heterogeneity and various owners of these resources, a seamless collaboration of the resources becomes harder to manage. Multi-agent technology could be used to meet this challenge, since agents are distributed in nature and autonomous and intelligent in behavior. Price is an important factor to motivate resource providers as well as users. One of the possible caveats to form an autonomous grid could be the dynamism of pricing mechanism. This paper focuses on two of the widely proposed economic models in this context; double auction and contract-netprotocol. We model a multiagent-based pricing frameworkwhere there are thousands of users with different resource demands as well as providers with their respective resource availability which are accommodated autonomously. A simulation environment is established and performances for different grid scenarios using the two different models are evaluated. We compare our results in terms of job rejection rate, total revenue gained by the provider and utilization of resources. The results show that in most cases, the double auction model performs better than contract-net-protocol. The contract-net model works well, when there are a few users and a few providers. Our findings could help grid resource providers to decide which model to use at a particular scenario in order to maximize profit.
A survey of economic models in grid computing
(Elsevier, 2011-04-16) Haque, Aminul; Alhashmi, Saadat M.; Parthiban, Rajendran
Grid computing offers the network of large scale computing resources. Economic models are effective in collaborating large scale heterogeneous grid resources that are typically owned by different organizations. Not all the models provide same benefits for users in utilizing the resources. Similarly, the profit earned by resource providers also differs for different economic models. We survey the economic models used in grid computing since its inception until 2010. We discuss their advantages and disadvantages and analyze their suitability for usage in a dynamic grid environment. To the best of our knowledge, no such survey has been conducted in the literature up to now. Highlights ► Economic approaches are efficient in Grid computing. ► Different economic models are proposed for Grid computing. ► Address strengths and weaknesses of different economic models. ► Different models are suitable for different scenarios. ► Proposal of using different models for different scenarios. Full Text Link: https://doi.org/10.1016/j.future.2011.04.009
A switching mechanism for grid providers
(IEEE Xplore, 2014-12-29) Haque, Aminul
Economic-based resource management models are found to be efficient for distributed resource collaboration in the Grid. Due to the wide range of applications and dynamic nature of Grid resources, constant performance by a specific economic model is therefore subjective. This work identifies the potential of different models in different scenarios in the Grid. This identification motivates us to design an optimization framework that couples suitable economic models and is able to switch from one model to another based on the models' domains of strength. The research further describes the roles played by an agent in dynamically deciding which model to be used when and for what purpose. It shows the effectiveness of the switching framework compared to any other individual models in a dynamic computing environment. Full Text Link: http://doi.org/10.1109/ICCITechn.2014.6997350
An Inspiration for Solving Grid Resource Management Problems Using Multiple Economic Models
(Springer, 2012) Haque, Aminul; Alhashmi, Saadat M.; Parthiban, Rajendran
Economic models can motivate resource providers to share resources across multiple administrations in Grid computing. Our survey on existing economic models in Grid computing identified that different economic models are suitable for different scenarios. In this paper, we conduct an experiment to quantify the strengths and weaknesses of widely proposed economic models in the Grid - Commodity Market, Continuous Double Auction, English Auction, Contract-Net-Protocol and Bargaining. Based on this experimental analysis, we identify regions where a particular economic model outperforms others. Then, we indicate that switching between the economic models could be used to maximize benefits in a specific scenario. Full Text Link: https://doi.org/10.1007/978-3-642-28675-9_1
An optimization-based adaptive resource management framework for economic Grids: A switching mechanism
(Elsevier, 2014-10-23) Haque, Aminul; Alhashmi, Saadat M.; Parthiban, Rajendran
The application of Grid computing has been broadening day by day. An increasing number of users has led to the requirement of a job scheduling process, which can benefit them through optimizing their utility functions. On the other hand, resource providers are exploring strategies suitable for economically efficient resource allocation so that they can maximize their profit through satisfying more users. In such a scenario, economic-based resource management strategies (economic models) have been found to be compelling to satisfy both communities. However, existing research has identified that different economic models are suitable for different scenarios in Grid computing. The Grid application and resource models are typically very dynamic, making it challenging for a particular model for delivering stable performance all the time. In this work, our focus is to develop an adaptive resource management architecture capable of dealing with multiple models based on the models’ domains of strengths (DOS). Our preliminary results show promising outcomes if we consider multiple models rather than relying on a single model throughout the life cycle of a Grid. Full Text Link: https://doi.org/10.1016/j.future.2014.10.022
Continuous Double Auction in Grid Computing: An Agent Based Approach to Maximize Profit for Providers
(IEEE Xplore, 2010-11-01) Haque, Aminul; Alhashmi, Saadat M.; Parthiban, Rajendran
Economic models are found efficient in managing heterogeneous computer resources such as storage, CPU and memory for grid computing. Commodity market, double auction and contract-net-protocol economic models have been widely discussed in the literature. These models are suitable for sharing distributed computer resources that belong to different owners. Agent technology can be used to manage these heterogeneous resources without human intervention, since agents are autonomous and intelligent in behavior. In this paper, we develop and simulate an agent-oriented double auction economic model. We compare the performance of our agent-oriented model with traditional double auction model, and show that the agent-oriented model is good in maximizing profit for providers. Full Text Link: http://doi.org/10.1109/WI-IAT.2010.105
Factors Causing Stunting Among Under-Five Children in Bangladesh
(Springer, 2020-10-22) Abid, Dm. Mehedi Hasan; Haque, Aminul; Hossain, Md. Kamrul
Malnutrition is one of the major problems in developing countries including Bangladesh. Stunting is a chronic malnutrition, which indicates low height for age and interrupt the growth. The purpose of this research is to find out the factors associated with the malnutrition status and test the accuracy of the algorithms used to identify the factors. Data from Bangladesh Demographic Health Survey (BDHS), 2014, is used. Factors like demographic, socioeconomic, and environmental have differential influence on stunting. Based on analysis, about 36% of under-five children were suffering from stunting. Decision tree algorithm was applied to find the associated factors with stunting. It is found that mothers’ education, birth order number, and economic status were associated with stunting. Support vector machine (SVM) and artificial neural network (ANN) are also applied with the stunting dataset to test the accuracy. The accuracy of decision tree is 74%, SVM is 76%, and ANN is 73%.
Five-Year Life Expectancy Prediction of Prostate Cancer Patients Using Machine Learning Algorithms
(Springer Nature Limited, 2022-03-08) Polash, Md Shohidul Islam; Hossen, Shazzad; Haque, Aminul
Prostate cancer is the most frequent malignancy and the leading cause of cancer-related mortality globally. A precise survival estimate is required for the effectiveness of treatment to minimize mortality rate. A remedial strategy can be planned under the anticipated survival state. Machine Learning (ML) methods have recently garnered considerable interest, particularly in developing data-driven prediction models. Unfortunately prostate cancer has received less attention to such studies. In this paper, we have built models using machine learning methods to predict whether a patient with prostate cancer would live for five years or not. Compared to prior studies, correlation analysis, a substantial quantity of data, and a unique track with hyperparameter adjustment boost the performance of our model. The SEER(Surveillance, Epidemiology, and End Results) database provided the data for developing these models. The SEER program gathers and disseminates cancer data to mitigate the disease’s effect. We analyzed prostate cancer patients’ five-year survival state using about seven prediction models. Gradient Boosting, Light Gradient Boosting Machine, and Ada Boost algorithms are identified as top-performed prediction models. Among them, a tuned prediction model using the Gradient Boosting algorithm outperforms others, with an accuracy of 88.45% and found fastest among the other models.
Identifying and Modeling the Strengths and Weaknesses of Major Economic Models in Grid Resource Management
(Springer, 2014-01-17) Haque, Aminul; Alhashmi, Saadat M.; Parthiban, Rajendran
Economic-based approaches have been found to be effective for distributed resource management in Grid computing. However, deciding which model to use is challenging, because (1) the performance stability of a particular model in a dynamic and distributed environment, is hard to establish (2) the performance objective of the Grid network may be complex, and it is difficult to know which model would best fit such an objective, (3) evidence indicates that no single model can cope with every scenario, and (4) no suitable tools exist to accurately predict and contrast the performances of one model with another model in a particular domain. Understanding the strengths and weaknesses of widely proposed economic models in terms of a range of scenarios is, therefore, crucial. To address this, the authors developed a general evaluation platform suitable for analyzing the performance of different economic models in the Grid. This work identifies domains of strength of individual models and highlights their use in various scenarios of Grid computing. Full Text Link: https://doi.org/10.1007/s10723-013-9289-9
Improved Vision-Based Diagnosis of Multi-Plant Disease Using an Ensemble of Deep Learning Methods
(Institute of Advanced Engineering and Science (IAES), 2023-10-15) Hridoy, Rashidul Hasan; Arni, Arindra Dey; Haque, Aminul
Farming and plants are crucial parts of the inward economy of a nation, which significantly boosts the economic growth of a country. Preserving plants from several disease infections at their early stage becomes cumbersome due to the absence of efficient diagnosis tools. Diverse difficulties lie in existing methods of plant disease recognition. As a result, developing a rapid and efficient multi-plant disease diagnosis system is a challenging task. At present, deep learning-based methods are frequently utilized for diagnosing plant diseases, which outperformed existing methods with higher efficiency. In order to investigate plant diseases more accurately, this article addresses an efficient hybrid approach using deep learning-based methods. Xception and ResNet50 models were applied for the classification of plant diseases, and these models were merged using the stacking ensemble learning technique to generate a hybrid model. A multi-plant dataset was created using leaf images of four plants: black gram, betel, Malabar spinach, and litchi, which contains nine classes and 44,972 images. Compared to existing individual convolutional neural networks (CNN) models, the proposed hybrid model is more feasible and effective, which acquired 99.20% accuracy. The outcomes and comparison with existing methods represent that the designed method can acquire competitive performance on the multi-plant disease diagnosis tasks.
Mn(III)-catalyzed Aerobic Oxidation of 3-Alkyl-4-hydroxy-1H-pyrrol-2 (5H)-ones in the Presence of 1,1-Diarylethenes. Synthesis of Stable 8-Aza1-hydroxy-2,3-dioxabicyclo[4.3.0]nonan-7-one Framework
(Wiley, 2014) Haque, Aminul; Nishino, Hiroshi
Twenty-two 3-alkyl-4-hydroxy-1H-pyrrol-2(5H)-ones were prepared and underwent Mn(III)-catalyzed aerobic oxidation in the presence of 1,1-diarylethenes to produce very stable crystalline 6-alkyl-8-aza-4, 4-diaryl-1-hydroxy-2,3-dioxabicyclo[4.3.0]nonan-7-ones in high yields. Full Text Link: http://doi.org/DOI 10.1002/jhet.1840
Model Analysis for Predicting Prostate Cancer Patient’s Survival: A SEER Case Study
(Springer Nature, 2023-05-28) Polash, Md. Shohidul Islam; Hossen, Shazzad; Haque, Aminul
Prostate cancer is assumed to be the most familiar cancer and the principal cause of death in the world. For effective treatment to decrease mortality, an accurate survival projection is essential. A remedy plan can be scheme under the predicted survival state. Machine Learning (ML) approaches have recently attracted significant attention, particularly in constructing data-driven prediction models. Prostate cancer survival prediction has received little attention in research. In this article, we constructed models with the support of ML techniques to determine the possibility of whether a patient with prostate cancer will survive or not. Feature impact analysis, a good amount of data, and a distinctive track make our model’s results better compared to previous research. The models have created using data from the SEER (Surveillance, Epidemiology, End Results) database. SEER program collects and distributes cancer statistics to lessen the disease impact. Using around twelve prediction models, we assessed the survival of prostate cancer patients. HGB, LGBM, XGBoost, Gradient Boosting, and Ada Boost are notable prediction models. Among them, the XGBoost is the best contribution, with an accuracy of 89.57%, and found to be faster among the models.
Ontology-based decentralized resource provisioning in economic grids
(Elsevier, 2013-03-26) Shaikh, Abdul Khalique; Haque, Aminul; Alhashmi, Saadat M.; Parthiban, Rajendran
A system that aggregates distributed and heterogeneous resources to solve computationally complex applications is known as Grid computing. One of the major challenges that current Grid systems are facing is the low resource utilization, resulting from the lack of suitable tools and mechanisms that understand the language of distributed applications. An effective utilization of resources depends on a better resource provisioning mechanism. The selection of resources in a Grid system involves finding and locating resources based on user requirements. Moreover, the performance of a Grid primarily depends on successful resource provisioning through scheduling and allocating resources according to users requirements. In this paper, we present a sub-domain ontology-based resource provisioning mechanism to increase the utilization of resources. We further extend the model to understand the characteristics of Grid entities in an economic system's point of view. We evaluate the significance of using dynamic pricing over static pricing to deal with the dynamic nature of the Grid. The results show improved success probability and system's profit compared to the traditional resource provisioning mechanisms.
PithaNet: A Transfer Learning-Based Approach for Traditional Pitha Classification
(Institute of Advanced Engineering and Science (IAES), 2023-10-05) Shakil, Shahriar; Akash, Atik Asif Khan; Nabi, Nusrat; Hasan, Mahmudul; Haque, Aminul
Pitha, pithe, or peetha are all Bangla words referring to a native and traditional food of Bangladesh as well as some areas of India, especially the parts of India where Bangla is the primary language. Numerous types of pithas exist in the culture and heritage of the Bengali and Bangladeshi people. Pithas are traditionally prepared and offered on important occasions in Bangladesh, such as welcoming a bride grooms, or bride, entertaining guests, or planning a special gathering of family, relatives, or friends. The traditional pitha celebration and pitha culture are no longer widely practiced in modern civilization. Consequently, the younger generation is unfamiliar with our traditional pitha culture. In this study, an effective pitha image classification system is introduced. convolutional neural network (CNN) pre-trained models EfficientNetB6, ResNet50, and VGG16 are used to classify the images of pitha. The dataset of traditional popular pithas is collected from different parts of Bangladesh. In this experiment, EfficientNetB6 and ResNet50 show nearly 90% accuracy. The best classification result was obtained using VGG16 with 92% accuracy. The main motive of this study is to revive the Bengali pitha tradition among young people and people worldwide, which will encourage many other researchers to pursue research in this domain.
Recognition of Bangladeshi Sign Language (BdSL) Words using Deep Convolutional Neural Networks (DCNNs)
(Ital Publications, 2023-12-01) Haque, Aminul; Pulok, Rishad Amin; Rahman, Md. Mizanur; Akter, Sanzida; Khan, Nusrat; Haque, Shamsul
In a world where effective communication is fundamental, individuals who are Deaf and Dumb (D&D) often face unique challenges due to their primary mode of communication—sign language. Despite the interpreters' invaluable roles, their lack of availability causes communication difficulties for the D&D individuals. This study explores whether the field of Human-Computer Interaction (HCI) could be a potential solution. The primary objective is to assist D&D individuals with computer applications that could act as mediators to bridge the communication gap between them and the wider hearing population. To ensure their independent communication, we propose an automated system that could detect specific Bangla Sign Language (BdSL) words, addressing a critical gap in the sign language detection and recognition literature. Our approach leverages deep learning and transfer learning principles to convert webcam-captured hand gestures into textual representations in real-time. The model's development and assessment rest upon 992 images created by the authors, categorized into ten distinct classes representing various BdSL words. Our findings show the DenseNet201 and ResNet50-V2 models achieve promising training and testing accuracies of 99% and 93%, respectively.
SkinNet-14: a deep learning framework for accurate skin cancer classification using low-resolution dermoscopy images with optimized training time
(Scopus, 2024-08-01) Mahmud, Abdullah Al; Azam, Sami; Khan, Inam Ullah; Montaha, Sidratul; Karim, Asif; Haque, Aminul; Hasan, Md. Zahid; Brady, Mark; Biswas, Ritu; Jonkman, Mirjam
The increasing incidence of skin cancer necessitates advancements in early detection methods, where deep learning can be beneficial. This study introduces SkinNet-14, a novel deep learning model designed to classify skin cancer types using low-resolution dermoscopy images. Unlike existing models that require high-resolution images and extensive training times, SkinNet-14 leverages a modified compact convolutional transformer (CCT) architecture to effectively process 32 × 32 pixel images, significantly reducing the computational load and training duration. The framework employs several image preprocessing and augmentation strategies to enhance input image quality and balance the dataset to address class imbalances in medical datasets. The model was tested on three distinct datasets—HAM10000, ISIC and PAD—demonstrating high performance with accuracies of 97.85%, 96.00% and 98.14%, respectively, while significantly reducing the training time to 2–8 s per epoch. Compared to traditional transfer learning models, SkinNet-14 not only improves accuracy but also ensures stability even with smaller training sets. This research addresses a critical gap in automated skin cancer detection, specifically in contexts with limited resources, and highlights the capabilities of transformer-based models that are efficient in medical image analysis.
Skinnet-14: A Deep Learning Framework for Accurate Skin Cancer Classification Using Low-resolution Dermoscopy Images with Optimized Training Time
(Springer, 2024-08-15) Mahmud, Abdullah Al; Azam, Sami; Khan, Inam Ullah; Montaha, Sidratul; Karim, Asif; Haque, Aminul; Hasan, Md. Zahid; Brady, Mark; Biswas, Ritu; Jonkman, Mirjam
The increasing incidence of skin cancer necessitates advancements in early detection methods, where deep learning can be beneficial. This study introduces SkinNet-14, a novel deep learning model designed to classify skin cancer types using low-resolution dermoscopy images. Unlike existing models that require high-resolution images and extensive training times, SkinNet-14 leverages a modified compact convolutional transformer (CCT) architecture to effectively process 32 × 32 pixel images, significantly reducing the computational load and training duration. The framework employs several image preprocessing and augmentation strategies to enhance input image quality and balance the dataset to address class imbalances in medical datasets. The model was tested on three distinct datasets—HAM10000, ISIC and PAD—demonstrating high performance with accuracies of 97.85%, 96.00% and 98.14%, respectively, while significantly reducing the training time to 2–8 s per epoch. Compared to traditional transfer learning models, SkinNet-14 not only improves accuracy but also ensures stability even with smaller training sets. This research addresses a critical gap in automated skin cancer detection, specifically in contexts with limited resources, and highlights the capabilities of transformer-based models that are efficient in medical image analysis.
Skinnet-14: A Deep Learning Framework for Accurate Skin Cancer Classification Using Low-resolution Dermoscopy Images with Optimized Training Time
(Springer Nature, 2024-08-01) Mahmud, Abdullah Al; Azam, Sami; Khan, Inam Ullah; Montaha, Sidratul; Karim, Asif; Haque, Aminul; Hasan, Md. Zahid; Brady, Mark; Biswas, Ritu; Jonkman, Mirjam
The increasing incidence of skin cancer necessitates advancements in early detection methods, where deep learning can be beneficial. This study introduces SkinNet-14, a novel deep learning model designed to classify skin cancer types using low-resolution dermoscopy images. Unlike existing models that require high-resolution images and extensive training times, SkinNet-14 leverages a modified compact convolutional transformer (CCT) architecture to effectively process 32 × 32 pixel images, significantly reducing the computational load and training duration. The framework employs several image preprocessing and augmentation strategies to enhance input image quality and balance the dataset to address class imbalances in medical datasets. The model was tested on three distinct datasets—HAM10000, ISIC and PAD—demonstrating high performance with accuracies of 97.85%, 96.00% and 98.14%, respectively, while significantly reducing the training time to 2–8 s per epoch. Compared to traditional transfer learning models, SkinNet-14 not only improves accuracy but also ensures stability even with smaller training sets. This research addresses a critical gap in automated skin cancer detection, specifically in contexts with limited resources, and highlights the capabilities of transformer-based models that are efficient in medical image analysis.
Survival Analysis of Thyroid Cancer Patients Using Machine Learning Algorithms
(IEEE, 2024-04-22) Alhashmi, Saadat M.; Polash, Md. Shohidul Islam; Haque, Aminul; Rabbe, Fazley; Hossen, Shazzad; Faruqui, Nuruzzaman; Hashem, Ibrahim Abaker Targio; Abubacker, Nirase Fathima
The medical community strives continually to improve the quality of care patients receive. Predictions of prognosis are essential for doctors and patients to choose a course of treatment. Recent years have witnessed the development of numerous new cancer survival prediction models. Most attempts to predict the prognosis of people with malignant growth rely on classification techniques. We could experiment with significantly different results using only a subset of SEER (Surveillance, Epidemiology, and End Results) data. These models were created using machine learning techniques by selecting univariate features and calculating correlations. We illustrated the variation in results and discrepancy of impurity that can result from varying data quantities and critical factors. Seventeen crucial factors were identified, and a group of classification algorithms were trained to evaluate the effectiveness of an estimation technique. In the display mode, the accuracy of these computations ranges from 97% to 99% A˙ long with accuracy, the models are further evaluated regarding the F1 score, precision, recall, and the AUC score. Compared to earlier studies, a more accurate model has been developed, and, to the best of our knowledge, our prediction model is superior to the models studied in the previous works.
Survival Analysis of Thyroid Cancer Patients Using Machine Learning Algorithms
(Scopus, 2024-04-22) Alhashmi, Saadat M.; Polash, Md. Shohidul Islam; Haque, Aminul; Rabbe, Fazley; Hossen, Shazzad; Faruqui, Nuruzzaman
The medical community strives continually to improve the quality of care patients receive. Predictions of prognosis are essential for doctors and patients to choose a course of treatment. Recent years have witnessed the development of numerous new cancer survival prediction models. Most attempts to predict the prognosis of people with malignant growth rely on classification techniques. We could experiment with significantly different results using only a subset of SEER (Surveillance, Epidemiology, and End Results) data. These models were created using machine learning techniques by selecting univariate features and calculating correlations. We illustrated the variation in results and discrepancy of impurity that can result from varying data quantities and critical factors. Seventeen crucial factors were identified, and a group of classification algorithms were trained to evaluate the effectiveness of an estimation technique. In the display mode, the accuracy of these computations ranges from 97% to 99% A˙ long with accuracy, the models are further evaluated regarding the F1 score, precision, recall, and the AUC score. Compared to earlier studies, a more accurate model has been developed, and, to the best of our knowledge, our prediction model is superior to the models studied in the previous works.