Muhammad Nazrul Islam

Ayon Roy, Tausif Al Zubayer, Nafisa Tabassum et al.

Regression analysis is a well known quantitative research method that primarily explores the relationship between one or more independent variables and a dependent variable. Conducting regression analysis manually on large datasets with multiple independent variables can be tedious. An automated system for regression analysis will be of great help for researchers as well as non-expert users. Thus, the objective of this research is to design and develop an automated curve fitting system. As outcome, a curve fitting system named "CurFi" was developed that uses linear regression models to fit a curve to a dataset and to find out the best fit model. The system facilitates to upload a dataset, split the dataset into training set and test set, select relevant features and label from the dataset; and the system will return the best fit linear regression model after training is completed. The developed tool would be a great resource for the users having limited technical knowledge who will also be able to find the best fit regression model for a dataset using the developed "CurFi" system.

Tareque Bashar Ovi, Shakil Mosharrof, Nomaiya Bashree et al.

The segmentation of satellite images is crucial in remote sensing applications. Existing methods face challenges in recognizing small-scale objects in satellite images for semantic segmentation primarily due to ignoring the low-level characteristics of the underlying network and due to containing distinct amounts of information by different feature maps. Thus, in this research, a tri-level attention-based DeepLabv3+ architecture (DeepTriNet) is proposed for the semantic segmentation of satellite images. The proposed hybrid method combines squeeze-and-excitation networks (SENets) and tri-level attention units (TAUs) with the vanilla DeepLabv3+ architecture, where the TAUs are used to bridge the semantic feature gap among encoders output and the SENets used to put more weight on relevant features. The proposed DeepTriNet finds which features are the more relevant and more generalized way by its self-supervision rather we annotate them. The study showed that the proposed DeepTriNet performs better than many conventional techniques with an accuracy of 98% and 77%, IoU 80% and 58%, precision 88% and 68%, and recall of 79% and 55% on the 4-class Land-Cover.ai dataset and the 15-class GID-2 dataset respectively. The proposed method will greatly contribute to natural resource management and change detection in rural and urban regions through efficient and semantic satellite image segmentation

Mohammad Amaz Uddin, Muhammad Nazrul Islam, Leandros Maglaras et al.

SMS, or short messaging service, is a widely used and cost-effective communication medium that has sadly turned into a haven for unwanted messages, commonly known as SMS spam. With the rapid adoption of smartphones and Internet connectivity, SMS spam has emerged as a prevalent threat. Spammers have taken notice of the significance of SMS for mobile phone users. Consequently, with the emergence of new cybersecurity threats, the number of SMS spam has expanded significantly in recent years. The unstructured format of SMS data creates significant challenges for SMS spam detection, making it more difficult to successfully fight spam attacks in the cybersecurity domain. In this work, we employ optimized and fine-tuned transformer-based Large Language Models (LLMs) to solve the problem of spam message detection. We use a benchmark SMS spam dataset for this spam detection and utilize several preprocessing techniques to get clean and noise-free data and solve the class imbalance problem using the text augmentation technique. The overall experiment showed that our optimized fine-tuned BERT (Bidirectional Encoder Representations from Transformers) variant model RoBERTa obtained high accuracy with 99.84\%. We also work with Explainable Artificial Intelligence (XAI) techniques to calculate the positive and negative coefficient scores which explore and explain the fine-tuned model transparency in this text-based spam SMS detection task. In addition, traditional Machine Learning (ML) models were also examined to compare their performance with the transformer-based models. This analysis describes how LLMs can make a good impact on complex textual-based spam data in the cybersecurity field.

Muhammad Nazrul Islam, Iyolita Islam, Kazi MD. Munim et al.

The objective of this research is to explore the existing mobile applications developed for the COVID-19 pandemic. To obtain this research objective, firstly the related applications were selected through the systematic search technique in the popular application stores. Secondly, data related to the app objectives, functionalities provided by the app, user ratings, and user reviews were extracted. Thirdly, the extracted data were analyzed through the affinity diagram, noticing-collecting-thinking, and descriptive analysis. As outcomes, the review provides a state-of-the-art view of mobile apps developed for COVID-19 by revealing nine functionalities or features. It revealed ten factors related to information systems design characteristics that can guide future app design. The review outcome highlights the need for new development and further refinement of the existing applications considering not only the revealed objectives and their associated functionalities, but also revealed design characteristics such as reliability, performance, usefulness, supportive, security, privacy, flexibility, responsiveness, ease of use, and cultural sensitivity.

Muhammad Nazrul Islam, Toki Tahmid Inan, Suzzana Rafi et al.

Artificial intelligence (AI) and machine learning (ML) have made a paradigm shift in health care which, eventually can be used for decision support and forecasting by exploring the medical data. Recent studies showed that AI and ML can be used to fight against the COVID-19 pandemic. Therefore, the objective of this review study is to summarize the recent AI and ML based studies that have focused to fight against COVID-19 pandemic. From an initial set of 634 articles, a total of 35 articles were finally selected through an extensive inclusion-exclusion process. In our review, we have explored the objectives/aims of the existing studies (i.e., the role of AI/ML in fighting COVID-19 pandemic); context of the study (i.e., study focused to a specific country-context or with a global perspective); type and volume of dataset; methodology, algorithms or techniques adopted in the prediction or diagnosis processes; and mapping the algorithms/techniques with the data type highlighting their prediction/classification accuracy. We particularly focused on the uses of AI/ML in analyzing the pandemic data in order to depict the most recent progress of AI for fighting against COVID-19 and pointed out the potential scope of further research.

Akib Zaman, Muhammad Nazrul Islam, Tarannum Zaki et al.

The objective of this article is to explore the Information and Communication Technology (ICT) interventions and its strengths, weaknesses, opportunities and threats for the containment of the pandemic spread of novel Coronavirus. The research adopted a qualitative research approach, while the study data were collected through online content review and Focus Group Discussion (FGD). Starting with a preliminary set of about 1200 electronic resources or contents, 56 were selected for review study, applying an inclusion and exclusion criteria. The review study revealed ICT interventions that include websites and dashboards, mobile applications, robotics and drones, artificial intelligence (AI), data analytic, wearable and sensor technology, social media and learning tools, and interactive voice response (IVR) as well as explored their respective usages to combat the pandemic spread of COVID-19. Later, the FGD was replicated with 22 participants and explored the possible strengths, weaknesses, opportunities, and threats (SWOT) of deploying such technologies to fight against the COVID-19 pandemic. This research not only explores the exiting status of ICT interventions to fight with the COVID-19 pandemic but also provides a number of implications for the government, practitioners, doctors, policymakers and researchers for the effective utilization of the existing ICT interventions and for the future potential research and technological development to the containment of the pandemic spread of COVID-19 and future pandemics.

Iyolita Islam, Kazi Md. Munim, Shahrima Jannat Oishwee et al.

Blockchain is relatively a new area of research. However, a surge of research studies on the blockchain has taken place in recent years. These research studies have mostly focused on designing and developing conceptual frameworks to build more reliable, transparent and efficient digital systems. While blockchain brings a wide variety of benefits, it also imposes certain challenges. Therefore, the objective of this research is to understand the properties of blockchain, its current uses, observed benefits and pitfalls to provide a balanced understanding of blockchain. A systematic literature review approach was adopted in this paper in order to attain the objective. A total of 51 articles were selected and reviewed. As outcomes, this research provides a summary of the state-of-the-art research studies conducted in the area of blockchain. Furthermore, we develop a set of concept maps aiming to provide in-depth knowledge on blockchain technology for its efficient and effective usage in the development of future technological solutions.