Shamiul Islam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shamiul Islam has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
Abu Baten
Northern University Bangladesh
4 shared publications- Imrus Salehin
Department of Computer Science and Engineering, Daffodil International University, Dhaka 1216, Bangladesh
3 shared publications Nazrul Amin
Northern University Bangladesh
3 shared publicationsJulkar Naeem
Bangladesh University of Business and Technology,Department of Computer Science and Engineering,Dhaka,Bangladesh,1216
2 shared publicationsTasmia Tahmida Jidney
Ahsanullah University of Science and Technology
2 shared publicationsUmmya Habiba
Bangladesh Agricultural University
2 shared publications
Latest Publications
- Open Access
- 19 November 2024
- Published by MDPI AG in Engineering Proceedings
- Vol. 76 (1) , 85
IoT-Based Smart Remote Door Lock and Monitoring System Using an Android Application
Nowadays, it is very important to secure our home perfectly. To make our life easier and more secure, we are presenting our smart door lock system project. We implement an IoT-based smart door lock system using an ESP32-CAM and an Android application in this project. Most of the time in our daily life, we forget to lock our doors and later we suffer from confusion about whether we locked all doors perfectly or not. In this project, we implement a smart door lock system, by which the owner can see the visitor’s picture and then lock or unlock their doors from anywhere and at any time using the Android application. Whenever visitors come to visit the home and press the doorbell, the owner will receive a notification on his/her smartphone and then the owner can see the visitor’s picture by using the Android app. After checking the visitor, the owner can let them enter the house by unlocking the door remotely. If the door is locked perfectly, then the door lock signal in the application will show a green signal. If the door is not locked perfectly, the signal will show red and then the owner can remotely lock their door easily from anywhere. In this project, we have also utilized a theft alert. If anyone comes in front of the door and tries to enter the house forcefully then a theft alert notification will be sent to the owner’s smartphone and a Buzzer Alert will ring in the house loudly so that the neighbors can be aware of the theft and can take action. The automatic door lock feature is also available in this system.
- 19 November 2024
- Published by MDPI AG in Engineering Proceedings
- Vol. 76 (1) , 85
Open AccessIoT-Based Smart Remote Door Lock and Monitoring System Using an Android Application
Nowadays, it is very important to secure our home perfectly. To make our life easier and more secure, we are presenting our smart door lock system project. We implement an IoT-based smart door lock system using an ESP32-CAM and an Android application in this project. Most of the time in our daily life, we forget to lock our doors and later we suffer from confusion about whether we locked all doors perfectly or not. In this project, we implement a smart door lock system, by which the owner can see the visitor’s picture and then lock or unlock their doors from anywhere and at any time using the Android application. Whenever visitors come to visit the home and press the doorbell, the owner will receive a notification on his/her smartphone and then the owner can see the visitor’s picture by using the Android app. After checking the visitor, the owner can let them enter the house by unlocking the door remotely. If the door is locked perfectly, then the door lock signal in the application will show a green signal. If the door is not locked perfectly, the signal will show red and then the owner can remotely lock their door easily from anywhere. In this project, we have also utilized a theft alert. If anyone comes in front of the door and tries to enter the house forcefully then a theft alert notification will be sent to the owner’s smartphone and a Buzzer Alert will ring in the house loudly so that the neighbors can be aware of the theft and can take action. The automatic door lock feature is also available in this system.
- 13 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
WebGuardML: Safeguarding Users with Malicious URL Detection Using Machine Learning
The prevalence of cyber security threats stemming from rogue URLs is widespread. A malicious URL contains phishing and spam to initiate assaults. Such websites attract innocent people who fall for frauds like identity theft and financial loss. Such threats must be identified and addressed promptly. The main goal of this research is to create a machine-learning-based malicious URL detection system. The method uses Extreme Gradient Boosting, Categorical Boosting, Decision Trees, Random Forest, Extra Trees, Gradient Boosting and others to detect hazardous URLs. The process begins with dataset collection, extracts domain attributes, path components, length features and addresses class imbalance. The best model is chosen after lengthy training and testing, considering accuracy, precision, recall and F1-score. Hyperparameter tuning and ensemble techniques are optional optimization procedures. The process of testing serves to assess the performance of a system in real-world scenarios. After deployment, the system is monitored and maintained to keep up with emerging dangerous URL trends. Overall, this research improves cyber security by automating URL protection.
- 13 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
WebGuardML: Safeguarding Users with Malicious URL Detection Using Machine Learning
The prevalence of cyber security threats stemming from rogue URLs is widespread. A malicious URL contains phishing and spam to initiate assaults. Such websites attract innocent people who fall for frauds like identity theft and financial loss. Such threats must be identified and addressed promptly. The main goal of this research is to create a machine-learning-based malicious URL detection system. The method uses Extreme Gradient Boosting, Categorical Boosting, Decision Trees, Random Forest, Extra Trees, Gradient Boosting and others to detect hazardous URLs. The process begins with dataset collection, extracts domain attributes, path components, length features and addresses class imbalance. The best model is chosen after lengthy training and testing, considering accuracy, precision, recall and F1-score. Hyperparameter tuning and ensemble techniques are optional optimization procedures. The process of testing serves to assess the performance of a system in real-world scenarios. After deployment, the system is monitored and maintained to keep up with emerging dangerous URL trends. Overall, this research improves cyber security by automating URL protection.
- Open Access
- 9 November 2023
- Published by Elsevier in SSRN Electronic Journal
Deep Learning Models for Real-Time Identification of Agro Plant Insects with Image Augmentation Effect Analysis on a Small Custom Dataset
An intelligent agricultural system is considered a crucial aspect of the modern world. Deep learning is the most popular and extensively tested approach to developing artificial intelligence. The development of agriculture through the application of deep learning techniques is a topic of significant interest. Deep learning models, especially those based on convolutional neural networks (CNNs), have been employed for precision agriculture. In our study, we investigate the impact of six distinct augmentation techniques on a limited dataset. We manually gathered data from the agricultural plant insect dataset (BAU-Insectv2) and employed four state-of-the-art CNN architectures (ResNet-101-v2, EfficientNet-B1, Inception v3 and Inception- ResNet v1 ) with various optimizers and augmentation techniques. We observed how these factorsinfluenced the dataset in our image classification experiments. Among the three optimizers (ADAM, SGD, RMS-prop), we achieved the best performance and highest accuracy with ADAM on a small dataset. We achieved superior image classification results when applying a minimum of four different augmentations and using an ADAM optimizer. Four CNN-based models are employed to enhance performance measures. The research focused on evaluating performance using various metrics, including accuracy, sensitivity, specificity, precision, recall, and F1- score. The primary aim of this study is to continually enhance the model’s performance over time. Finally, we have presented a comprehensive framework designed for real-time crop image classification which includes the development of a user-friendly web application for seamlessand efficient operation.
- 9 November 2023
- Published by Elsevier in SSRN Electronic Journal
Open AccessDeep Learning Models for Real-Time Identification of Agro Plant Insects with Image Augmentation Effect Analysis on a Small Custom Dataset
An intelligent agricultural system is considered a crucial aspect of the modern world. Deep learning is the most popular and extensively tested approach to developing artificial intelligence. The development of agriculture through the application of deep learning techniques is a topic of significant interest. Deep learning models, especially those based on convolutional neural networks (CNNs), have been employed for precision agriculture. In our study, we investigate the impact of six distinct augmentation techniques on a limited dataset. We manually gathered data from the agricultural plant insect dataset (BAU-Insectv2) and employed four state-of-the-art CNN architectures (ResNet-101-v2, EfficientNet-B1, Inception v3 and Inception- ResNet v1 ) with various optimizers and augmentation techniques. We observed how these factorsinfluenced the dataset in our image classification experiments. Among the three optimizers (ADAM, SGD, RMS-prop), we achieved the best performance and highest accuracy with ADAM on a small dataset. We achieved superior image classification results when applying a minimum of four different augmentations and using an ADAM optimizer. Four CNN-based models are employed to enhance performance measures. The research focused on evaluating performance using various metrics, including accuracy, sensitivity, specificity, precision, recall, and F1- score. The primary aim of this study is to continually enhance the model’s performance over time. Finally, we have presented a comprehensive framework designed for real-time crop image classification which includes the development of a user-friendly web application for seamlessand efficient operation.
- Top Cited PapersOpen Access
- 18 October 2023
- Published by Elsevier in Journal of Information and Intelligence
- Vol. 2 (1) , 52-81
AutoML: A systematic review on automated machine learning with neural architecture search
- 18 October 2023
- Published by Elsevier in Journal of Information and Intelligence
- Vol. 2 (1) , 52-81
Top Cited PapersOpen AccessAutoML: A systematic review on automated machine learning with neural architecture search
- 16 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Prediction of Buying Intention: Factors Affecting Online Shopping
Recent times have witnessed a very significant increase in the use of electronic commerce all over the world. The fact that sales have been steadily climbing demonstrates that there is massive untapped market potential for online shopping. The process of analysing data to determine and categorise the intentions of online shoppers contributes to the accumulation of store earnings. The application of machine learning classification models to the data collected from e-commerce websites is the main goal of our endeavours. Specifically, We investigate whether machine learning is a reliable method for predicting the possibility that a consumer who explores a retailer's website will actually make a purchase. Exploratory data analysis was carried out so that we could visually examine our information and search for any patterns or trends that may have been there. In this article, we present a method that is capable of predicting the purpose of consumers who visit an e-commerce site to make a purchase based on data obtained while these users are browsing the websites in question. The Online Consumer Buying Sentiment Dataset that can be found in the UCI Machine Learning Repository is one that we use.
- 16 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Prediction of Buying Intention: Factors Affecting Online Shopping
Recent times have witnessed a very significant increase in the use of electronic commerce all over the world. The fact that sales have been steadily climbing demonstrates that there is massive untapped market potential for online shopping. The process of analysing data to determine and categorise the intentions of online shoppers contributes to the accumulation of store earnings. The application of machine learning classification models to the data collected from e-commerce websites is the main goal of our endeavours. Specifically, We investigate whether machine learning is a reliable method for predicting the possibility that a consumer who explores a retailer's website will actually make a purchase. Exploratory data analysis was carried out so that we could visually examine our information and search for any patterns or trends that may have been there. In this article, we present a method that is capable of predicting the purpose of consumers who visit an e-commerce site to make a purchase based on data obtained while these users are browsing the websites in question. The Online Consumer Buying Sentiment Dataset that can be found in the UCI Machine Learning Repository is one that we use.
- 15 February 2023
- Published by Springer Nature
- p. 258-265
Improve CNN Model Agro-Crop Leaf Disease Identification Based on Transfer Learning
- 15 February 2023
- Published by Springer Nature
- p. 258-265
Improve CNN Model Agro-Crop Leaf Disease Identification Based on Transfer Learning
- 29 January 2023
- Published by Springer Nature
- p. 209-217
Hybrid Convolution Neural Network with Transfer Learning Approach for Agro-Crop Leaf Disease Identification
- 29 January 2023
- Published by Springer Nature
- p. 209-217
Hybrid Convolution Neural Network with Transfer Learning Approach for Agro-Crop Leaf Disease Identification
- 26 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Structural Analysis of the Suction Valve Leaf of Household Refrigerator Compressors: Effects of Edge Radius
The performance of household refrigerator compressor depends mainly on the mass flow rate through the reed valves. The suction and discharge valve leaves are designed to trap and deliver the refrigerant respectively during each piston movement cycle. Therefore, the edge radius of the valve leaves is a vital parameter that affects the mass flow rate, structural performance, and the fatigue life. In this research, the effects of different edge radii on the suction valve leaf for different materials have been determined with static structural analysis. Additionally, the natural frequency has been determined with modal analysis for the free and undamped cases to establish the relationship with different edge radii. The numerical simulation proved that the factor of safety increase with the increment of edge radius but there is no significant effect on vibration. Also, the fatigue life has been determined.
- 26 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Structural Analysis of the Suction Valve Leaf of Household Refrigerator Compressors: Effects of Edge Radius
The performance of household refrigerator compressor depends mainly on the mass flow rate through the reed valves. The suction and discharge valve leaves are designed to trap and deliver the refrigerant respectively during each piston movement cycle. Therefore, the edge radius of the valve leaves is a vital parameter that affects the mass flow rate, structural performance, and the fatigue life. In this research, the effects of different edge radii on the suction valve leaf for different materials have been determined with static structural analysis. Additionally, the natural frequency has been determined with modal analysis for the free and undamped cases to establish the relationship with different edge radii. The numerical simulation proved that the factor of safety increase with the increment of edge radius but there is no significant effect on vibration. Also, the fatigue life has been determined.