This study addresses the challenges of predicting user preferences for songs by utilizing machine learning algorithms. Existing research in this area has primarily focused on user-based collaborative filtering or content-based approaches, neglecting the potential of utilizing song attributes for personalized song recommendations. Several algorithms are evaluated in this study, including Random Forest Classifier, Logistic Regression, Gaussian Naive Bayes, Extreme Gradient Boosting, Dummy Classifier, and Stacking Classifier. The Stacking Classifier model was chosen as the best model due to its consistently high accuracy, precision, recall, and F1 score. Spotify API is used in the deployment process to retrieve song attributes, encode them, and input them into the model for prediction. In addition, the model's accuracy is evaluated using two different playlists, with predicted results of songs that the user would like or dislike. Overall, the study suggests that the Stacking Classifier model is suitable for predicting song preferences on Spotify. Furthermore, the deployment process outlined in this study offers a convenient tool for users to predict their preferences for individual songs or playlists. This can empower users to curate their music collections more effectively and help music streaming platforms like Spotify to further improve their recommendation systems.
Wireless Sensor Networks (WSNs) that monitor individuals' physical and medical conditions are called Body Area Networks (BANs). This technology uses wireless sensors to monitor vital signs from the human body. In order to overcome processing and storage limitations, the BAN is often connected to the cloud. Wireless Body Area Network (WBAN), however, opens up the possibility of various security threats when using the cloud. The digital twin (DT) can provide users with hundreds of thousands of interpretations of the physical device without interfering with its normal functionality. It provides an accurate depiction of a cyberattack before time, which maximizes efficiency by reflecting and simulating the physical devices and their relationships with the environment. This paper proposes an integrated security framework for cloud-assisted BANs (CBANs), which will use DT to detect cyberattacks. By combining digital and physical models, cyber-physical security (CPS) can be optimized in advance at low risk and cost. The results of the study show that designing and evaluating techniques for the protection of CBAN using digital twin will help security experts to use an optimized solution without physical testing that will also save time and resources.
Haze is an atmospheric phenomenon that occurs mostly in developing countries and is caused by tiny micro-gaseous air pollutants that affect human health, the economy, and the environment. The transboundary haze, or polluted air scattered in the atmosphere from a source location, impacts the livelihoods of the human population, lasting days to weeks. To prevent massive disruption caused by haze, it is thus important to detect the probable affected locations by knowing the source location of the pollutants in the atmosphere. Based on the idea of the Epidemiological Triangle (ET), we proposed a contact triangle named Haze Contact Triangle (HCT) be formulated, representing the interdependency of the environment, source locations, and affected locations. Therefore, this paper aims to present a proof-of-concept that in detecting the affected locations, a bipartite network model can be used for this purpose. The relationship between two nodes which are the source location and the affected location of the haze was clearly defined and quantified through a weighted link. The quantification of the two nodes is based on the environmental properties of the locations so that the dispersion of haze from the source location to the affected location can be formulated. The weight of the link between the two nodes was quantified using parameters such as the high temperature, wind directions, distance between the affected location from the source location, and the rates of smouldering fire in the formulation of the Haze Contact Strength (HCS). The formulated network is then sent through a web-based searching algorithm resulting in ranked locations. Therefore, the result is an indication of a haze hotspot, which may then be used to improve the efficacy of controlling the haze situation locally.
Fog computing is paving the way for catering latency stringent applications (e.g., augmented reality and virtual reality) at the customer premises. Fog computing nodes are located between the end user devices and remote cloud. They are lightweight and small-scale storage and processing system deployed closer to the data source, allowing faster processing as well as providing privacy and security of the data. We have been witnessing a growing number of solutions that integrate Passive Optical Network (PON) with the fog computing. Along this research line, there are some solutions considering fog computing nodes are co-located with the Optical Line Terminal (OLT), the central intelligence of a PON system, and Optical Network Unit (ONU), a customer premises equipment. There are some solutions, on the other hand, that propose to embed computing and storage functionality within PON equipment itself. In this paper, we particularly focus on the later approach. Here, we propose an energy conserving solution for a PON system with fog computing enabled ONUs, i.e., the ONUs are equipped with additional Computing and Storage Units (CSUs). Our solution aims at minimizing energy consumption of a PON system by keeping only the required number of CSUs active and allowing the ONUs to move into sleep mode (whenever possible) by taking into account task arrival rate and task completion deadline.
The Shor algorithm demonstrates the significant risk that quantum attacks pose to the security of widely used cryptographic primitives. However, code-based cryptography has been shown to be resistant to these attacks. To date, no polynomial-time attack exists that can break code-based cryptosystems such as the McEliece cryptosystem. Despite this, these cryptosystems are not employed in practical applications in domains such as online banking, blockchains, and e-commerce platforms. The primary reason is the large sizes of the public and private keys associated with code-based cryptosystems. In this paper, a new code-based cryptosystem is introduced which employs a dual matrix based on the transpose and inverse of the parity check matrix to reduce the key size compared to the McEliece cryptosystem.
Colorectal Cancer (CRC) is a prevalent and deadly disease, and accurate and timely diagnosis is essential for improving patient outcomes. The use of deep learning in medical imaging offers a promising avenue for achieving this goal. The ability to accurately identify different types of cancer cells can aid in treatment planning and prognosis and may ultimately help to save lives. This study proposes two models, radiomic-based Support Vector Machine (SVM) and a deep-learning model to recognize different types of cells in colorectal cancer using pathological images. In the first model, the radiomics features are extracted from the histopathology images and SVM used for CRC classification. The second model extracted the deep learning features and classified the CRC using Res-Net-18. The study utilized a dataset of 5000 pathological images of colorectal cancer, with eight classes of cells to be recognized. The deep learning model achieved high scores in terms of recall, precision, F1-score, and accuracy for each class, with an overall accuracy score of 0.95. These results demonstrate the potential of deep learning in medical imaging and cancer diagnosis. Our findings suggest that deep learning could be a powerful tool for accurately diagnosing different types of cancer cells, aiding in treatment planning and prognosis. Finally, this study contributes to the growing body of literature on the use of deep learning in medical imaging and cancer diagnosis.
In the current era where Augmented Reality (AR) is making significant impact on various aspects of education, its specific role in mathematics education is of crucial significance. A quantitative research design was employed, utilizing the Fuzzy Delphi technique to collate opinions from a diverse group of 17 experts drawn from fields such as education, User Interface/User Experience (UI/UX), mobile app development, AR/VR, mathematics, computer science, and educational technology. The data analysis indicated a consensus among the experts on the key elements necessary for creating effective AR applications in mathematics education, as evidenced by values exceeding 75% agreement, with a maximum acceptable dispersion of opinions (d) ≤ 0.2, and a minimum consensus measure (fuzzy score, ≥ α-cut value larger than 0.5). The findings underscore the importance of the identified components in the design and development of AR applications for mathematics education and signal significant potential for promoting technology-enhanced active learning in the educational field.
Small and medium-sized businesses or known as SMEs are essential contributors in the software industry. Software Requirement Engineering (SRE) process is an essential part of the software development life cycle. With the advent of technology, the software has become necessary component of business, especially with Small and Medium Enterprises (SMEs). Due to this many challenges faced by IT SMEs in adoption of SRE for successful completion of projects. There have been many factors affecting the progress of SMEs among which few are reported in this SLR. A total of 40 studies are found which are relevant to the research questions and published between 2010 and 2023. The research highlights that SMEs face significant challenges when implementing effective SRE processes due to a lack of resources, experience, and tools. However, the studies propose various solutions to address these challenges, including Agile methodologies, communication enhancement, collaboration among stakeholders, and tools to support SRE processes. The article emphasizes the importance of considering the human elements of SRE, such as team dynamics, training, and motivation, and suggests that SMEs should establish specialized SRE processes that fit their requirements and resources. The article concludes that more research is needed to investigate the effectiveness of SRE practices in SMEs and to provide guidelines and frameworks for improving SRE processes in this significant segment of the software industry. Overall, the research provides helpful insights into the current state of SRE practices in SMEs and highlights the potential for enhancing SRE procedures in this sector.
Construction industry significantly contributing to nation’s economy, but recent studies identified the industry having -1.4 rate of productivity deterioration. Research studied that every decrease of 10% workforce performance results to 5% lower profitability, profitability is critical for organization survival. Systematic technology implementation is the key of grooming performance up to 260%. The paper aimed to explore a systematic technology implementation framework to support the industry for effective digital application and enhance work performance. With that, the paper combines the use of two literature review techniques, which are Systematic Literature Review (SLR) technique and Integrative Literature Review (ILR) technique. SLR technique took place to study the technology application components and ILR technique propose the systematic technology implementation framework based-on literature cases. Though SLR studies found that varied research concluded diverse technology implementation components, are pointing into three main components categories, namely digital data, internet connectivity, and intelligent enablement. ILR technique further navigates the three component categories to form as a systematic technology implementation framework for effective technology implementation. As the discussion section the paper indicated that, systematic technology implementation framework forms by first having digital data, then connects with internet connectivity, and process with intelligent enablement. It helps to support the industry for effective technology implementation and improve work performance. It aids practitioners adopt technology effectively to improve work performance and avoid suffering from loss of investment and resistance of transf ormation.
This paper presented a framework of data quality assurance by applying machine learning. Data quality assurance often a major concern in the decision-making derivation since decades. Poor quality of the data will eventually affect the result of the decision making. Despite the frequent occurrence of missing data, reliability of data and inconsistency of data, there are a few machine learning algorithms that can perform prevention of poor-quality data at early stage of the data acquisition and data cleansing effectively. This research aims to solve the challenges mentioned above through the framework designed. The framework is designed to integrate two algorithms into the data management lifecycle. Preventive algorithm meant to be applied in the earlier stage of the data management lifecycle while predictive algorithm is applied for the purpose of data cleansing. It is foresee that this framework is able to improve the data quality. Future work will be the implementation of the framework.
Over the years, flood and drought events are some of the common primary impacts of climate change. Local water experts believe that water-related risks are often caused by prolonged droughts and flash floods in Brunei Darussalam. Besides these two occurrences, the water quality of the river is also drastically affected. This study investigates the feasibility of implementing a real-time monitoring system of water quality in the river. It aims to confront one of the century-long problems of flood and address the lack of a real-time warning system. The system utilizes a cost-effective and efficient monitoring system that consists of identified water quality sensors to be used that can automatically detect and act as an early warning and notify the Department of Water Services (DWS) for any form of abnormalities in water quality. The system will be useful for the authorities in the Water Treatment Plant as it will provide useful information that can help them in performing the necessary treatment accordingly, especially in the event of heavy rainfall. The use of IoT allows the connections among devices to gather and transmit data remotely in the defined environmental applications.
Computed Tomography (CT) imaging has become a commonly used technique in healthcare to identify irregularities in the human body. However, CT scans involve exposure to electromagnetic radiation, which can pose health risks to patients. To address this, Low-Dose CT has been introduced, but it results in degraded image quality, including increased noise, artifacts, and loss of edge and feature contrast. This can limit the effectiveness of Computer-Aided Diagnosis systems. Denoising and preserving edge sharpness in Low-Dose CT images is a challenging task that conventional denoising techniques may not efficiently solve. Deep learning-based methods have emerged as a potential solution to this problem. This study proposes a new unsupervised LDCT image denoising algorithm called DEPnet (Denoise and Edge Preserve), which utilises a U-Net-based autoencoder with hybrid dilated convolution and batch normalization layers. The proposed algorithm has been evaluated on the KiTS19 Low-Dose CT Grand Challenge dataset and compared with other models such as Q-AE, Msaru-Net, and CT-ReCNN. The results demonstrate that DEPnet effectively reduces noise in LDCT images and preserves fine details, making it a promising solution for denoising Low-Dose CT images.
The internet of things (IoT) has been embedded in many aspects of our lives and is evolving in almost all sectors such as (smart homes, smart cities, smart hospitals, etc.). As security is the main concern everywhere in daily routine due to the increase in cases of crimes and vulnerabilities. Intrusion detection systems (IDS) are extremely important to make the specific location secure from unauthorized access but still the perimeter intrusion detection systems (PIDS) are facing issues in real intrusion detection and false alarm rate (FAR) that are caused by the environmental intrusion. In order to solve these problems in perimeter intrusion detection systems (PIDS). In this paper, we proposed a new machine learning-based model named STPID-Model where we used the Imagery library for intelligent detection systems (i-LIDS) dataset, we derived images from the recordings available in i-LIDS dataset and then we applied enhanced algorithm I-DBSCAN for intrusion detection and distinguish between fake and real intrusion. Our proposed model performed better than the others.
Technology has increased the interest and demand for pervasive systems which require contextual information to function at optimal capacity. There have been numbers of research in context-aware systems that has limited focused on the semantic-based approach in the crowdsourcing domain. Thus, it promotes challenges in the context and service acquisition and representation for reasoning control mechanism. This paper aims to review semantic-based reasoning framework with a focus on the mobile crowdsourcing domain. Different domains acquire different contextual information, either extrinsic or intrinsic. The review of the frameworks has help to formulate a process framework that applied semantic approach that has the important component for context-aware reasoning process. The framework can be used in the context-aware mobile crowdsourcing domain or can be generalized to other domain to aid reasoning control. Its advantage over other crowdsourcing frameworks which is focuses not only on context and service acquisition but also the representation on the acquired information.
With the advent of Advanced Persistent Threats (APTs), it has become more challenging to effectively detect and comprehend computer system attacks. This paper proposed an Intrusion Detection System (IDS) to effectively detect APT activities in each stage of the APT life cycle using decision trees and gradient-boosting algorithms. In addition, this model generates APT fingerprints by optimizing APT stages or attack paths that help the model with early APT detection. This model is evaluated and validated using Dataset APT (DAPT) 2020. The proposed model proved that effectively classified APT activities with more than 97.63 accuracy in most APT stages. Furthermore, this model proved effective in generating APT fingerprints.
Accurate sentiment analysis is greatly hindered by the code-switching phenomena, especially in the setting low resource language such as the Hausa. However, the majority of previous studies on Hausa sentiment analysis have mainly ignored this problem. This study explores the use of transformer fine-tuning techniques for Hausa language sentiment classification tasks using three pre-trained multilingual language models: Roberta, XLM-R, and mBERT. A multilabel sentiment classification was conducted using Python programming language and TensorFlow library, with a GPU hardware accelerator on Google Collaboratory. The Twitter dataset used in this study contains 16849 train and 2677 unlabeled dev and 5303 test unlabeled samples of tweets/accounts, each labelled with positive, negative, and neutral respectively for train set data. The findings demonstrate that the mBERT-base-cased model gets the maximum accuracy and F1-score of 0.73% and 0.73%, respectively, outperforming the other two pre-trained models. The train and validation accuracy graph of the mBERT model shows improvement over time. The study underscores the importance of tailoring the implementation code to meet specific requirements and the significance of fine-tuning pre-trained models for optimal performance.
Aquaculture is one of the emerging industries capable of closing the demand and supply gap for aquatic products. However, the current method of farming is facing sustainability challenges due to its resource-intensive nature, particularly in terms of water and feed usage. Additionally, the rapid growth of the aquaculture sector has led to issues such as overfeeding. The Department of Fisheries Brunei Darussalam has introduced several initiatives to boost the output of aquaculture. With assistance from the Department of Fisheries and Hiseaton Fisheries Ltd, this study aims to implement an IoT-based system to monitor water quality to culture marine species, in particular Sea Bass & Giant Freshwater Prawn, a popular choice for aquaculture in Asia. These species have a high environmental tolerance for thriving in both marine and freshwater. Traditional water monitoring has been a standard way to measure water quality in aquaculture industries in Brunei Darussalam. Water quality is the most important factor that determines the growth rate and health of aquatic creatures. The proposed setup employs a Recirculating Aquaculture System together with the Internet of Things in the rearing of Sea Bass and Giant Freshwater prawns by monitoring the water quality parameters, automating the feeding schedule, and regulating the water quality.
Wireless Technology is developing very fast. Researchers are actively researching in wireless communication as the technology for wireless communication has been growing quickly. Vehicular Ad Hoc Networks (VANETs), a cutting-edge technology in this area, have the potential to make a significant contribution to smart transportation systems in the future. VANETs offer a framework for communication that enhances traffic services and aids in lowering accident rates. Maintaining stability in Vehicular Ad-Hoc Network (VANET) clustering is difficult tasks due to high node mobility. First issue in VANET clustering is the Cluster Head (CH) selection since the CH has critical role in data routing and responsible for coordinating both inter and intra cluster communication. Second issue is the high mobility of nodes that cause difficulty to retain clustering optimization and will lead to inefficiency in network communication. Introduce MFO algorithm for simulate the movement behavior of moths and update the position based upon movements. Proven to be an effective and efficient method for solving optimization problem. To design K-Means algorithm that portion nodes based on their proximities by optimize the distance between nodes within same cluster by assigning them to the closet cluster center. Improving clustering efficiency by sending frequent updates to the CH in term of improving scalability, coverage, and clustering result, while reducing communication and energy consumption. Overall, the MFO Algorithm and K-Means algorithm can be used in combination to optimize the clustering in VANET, leading to better network performance, more reliable communication, and improved efficiency.
In response to the COVID-19 pandemic, the advancement of automated services in the financial technology (Fin-tech) sector has undeniably sped up tremendously. At the same time, the instability of the financial market and inflation have influenced specific groups of potential and existing investors to reconsider investment choices and re-manage their financial planning. In addition, a few studies are investigating the financial literacy, digital literacy, information literacy towards intention to use Robo-advisors. For this reason, this paper aims to provide a concise review on the behaviors towards investment robo-advisors which includes discussions on Fin-tech, types of robo-advisors, history, features, and technological perspectives. This paper can assist the system developers in various industries.
The CSMS/CA protocol is employed in wireless networks in order to overcome issues such as the hidden node problem. This mechanism is expected to handle collisions better using the RTS/CTS mechanism. This method will allow a participating node to take part in communication only if it receives a “Clear to Send” message and thereby, theoretically “avoiding” collision. The objective of this paper is to analyse the improvement that the RTS/CTS mode brings over the Basic Access mode. The paper presents the study of wireless nodes within a specific area with increasing node concentration to verify the performance impact of a protocol in wireless networks, particularly when the node concentration increases.
