The rise of online education has changed the way students usually learn by making educational materials easier to get to and creating a global learning community. While online education offers numerous benefits, it is also crucial to acknowledge its certain drawbacks, such as the potential reduction in interaction between students and teachers, which might increase signs of isolation among students and impede opportunities for collaborative learning. Therefore, Student Evaluations of Teaching (SET) play a critical role in identifying areas for improvement from the students' standpoint, thereby promoting constructive communication between students and teachers. This research conducts a comparison among the traditional Educational Data Mining (EDM) techniques to find out the best-performing classifier for analyzing student evaluations of teaching online. It is accomplished by first extracting the dataset from the student evaluations of teaching at X-University and then applying six different classifiers to the dataset that were extracted. The results demonstrated that Logistic Regression, Naive Bayes, and K-Nearest Neighbors (KNN) exhibited a notably high level of accuracy compared to other classification techniques. The findings of this research will provide guidance for future researchers in applying a wider range of classification techniques to extensive datasets and in implementing the necessary adjustments to achieve superior results.

With the rapid proliferation of drones and drone networks across various application domains, ensuring their security against cyber threats has become imperative. This paper presents a comprehensive analysis and comparative analysis of the state-of-the-art techniques for detecting cyber threats in drone networks. The background provides a primer on drones, networks, drone network architectures, communication mechanisms, and enabling technologies like wireless protocols, satellite navigation, onboard computers, sensors, and flight control systems. The landscape of emerging technologies including blockchain, software-defined networking, machine learning, fog computing, ad-hoc networks, and swarm intelligence is reviewed in the context of transforming drone network capabilities while also introducing potential vulnerabilities. The paper delves into common cyber threats faced by drone networks such as hacking, DoS attacks, data breaches, and GPS spoofing. A detailed literature review of proposed threat detection techniques is provided, categorized into machine learning, multi-agent systems, blockchain, intrusion detection systems, software solutions, and miscellaneous methods. A key gap identified is handling increasingly sophisticated attacks, complex environments, and resource limitations in aerial platforms. The analysis highlights accuracy, overhead and real-time trade-offs between techniques, while factors like model optimization can influence efficacy. A comparative analysis highlights the advantages and limitations of each approach considering metrics like accuracy, scalability, flexibility, and overhead. Key observations include the trade-offs between computational complexity and real-time performance, the challenges in handling evolving attack techniques, and the dependencies between detection accuracy and factors like model selection and training data quality. The analysis provides a comprehensive reference for cyber threat detection in drone networks, benefiting researchers and practitioners aiming to advance this crucial area of drone security through robust detection systems tailored for resource-constrained aerial environments.

Systems for registering vehicles are essential for keeping track of ownership changes. However, severe flaws in the current systems permit vehicles that have been stolen or illegally sold to be registered. Inefficient verification techniques, drawn-out administrative processes, and dishonest employees cause these problems. This paper introduces a transparent system to prevent denial, alteration, or unauthorized manipulation. The proposed method employs hybrid blockchain architecture, distinguishing between confidential and non-confidential data. Personal information is stored privately, while vehicle-related data is maintained as public information. The adoption of blockchain technology is driven by its robust security features, transparency, and traceability, as well as its immutability and ability to handle many users effectively.

An Enterprise Resource Planning (ERP) system is a software application that serves as a centralized platform to streamline and automate organizational functions and share real-time data, facilitating efficient communication and collaboration. It provides an all-inclusive approach to managing and optimizing business processes, boosting efficiency, fostering cooperation, and giving an overall picture of how the organization is operating. However, the traditional centralized databases in ERP systems pose security concerns. Blockchain Technology can be an appealing alternative as it comes with immutable and decentralized data as well as enhanced security. This study focuses on two methods of securing data management in ERP systems: Organizing the distributed information using The Ralph Kimball data model and optimizing an individual block using Database Sharding. This study does an extensive examination to determine the effectiveness of both suggested strategies, comprising a detailed evaluation that highlights the benefits and limitations of both techniques. This paper intends to patch the security holes in ERP systems to safeguard sensitive data and mitigate risks.

The process of making decisions on software architecture is the greatest significance for the achievement of a software system's success. Software architecture establishes the framework of the system, specifies its characteristics, and has significant and major effects across the whole life cycle of the system. The complicated characteristics of the software development context and the significance of the problem have caused the research community to build various methodologies focused on supporting software architects to improve their decision-making abilities. With these efforts, the implementation of such systematic methodologies looks to be somewhat constrained in practical application. Moreover, the decision-makers must overcome unexpected difficulties due to the varying software development processes that propose distinct approaches for architecture design. The understanding of these design approaches helps to develop the architectural design framework. In the area of software architecture, a significant change has occurred wherein the focus has shifted from primarily identifying the result of the architecting process, which was primarily expressed through the representation of components and connectors, to the documentation of architectural design decisions and the underlying reasoning behind them. This shift finally concludes in the creation of an architectural design framework. So, a correct decision- making approach is needed to design the software architecture. The present study analyzes the design decisions and proposes a new design decision model for the software architecture. This study introduces a new approach to the decision-making model, wherein software architecture design is viewed based on specific decisions.

This paper explores the proposed solutions based on blockchain technology's potential to solve supply chain management issues. The problems include lack of traceability and transparency, scalability and cost issues, sustainability, efficiency, patchwork logistics, and bullwhip effect issues. In this paper, we have suggested some solutions with the help of blockchain technology. The solutions can solve multiple significant issues in supply chain management. Our blockchain-based solutions can provide a secure and visible record of all transactions and data along the supply chain, which can improve traceability and transparency, a decentralized and efficient method of data processing and exchange that can also increase scalability and reduce cost, a transparent and accountable way to track and verify sustainability-related data. Our method can enable more streamlined and automated tracking and data sharing, helping to reduce the risk of delays and inefficiencies while mitigating the risk of the bullwhip effect by providing real-time visibility and enabling better communication and collaboration between parties. The paper discusses the implications and challenges of implementing blockchain in supply chain management.

Heart disease is very common in today’s day and age, with death rates climbing up the numbers every year. Prediction of heart disease cases is a topic that has been around in the world of data and medical science for many years. The study conducted in this paper makes comparison of the different algorithms that have been used in pattern analysis and prediction of heart diseases. Among the algorithms that have been used in the past included a combination of machine learning and data mining concepts that essentially are derived from statistical analysis and relevant approaches. There are a lot of factors that can be considered when attempting to analytically predict instances of heart diseases, such as age, gender, resting blood pressure etc. Eight such factors have been taken into consideration for carrying out this qualitative comparison. As this study uses a particular data set for extracting results from, the output may vary when implemented over different data sets. The research includes comparisons of Naive Bayes, Decision Tree, Random Forest and Logistic Regression. After multiple implementations, the accuracy in training and testing are obtained and listed down. The observations from implementation of these algorithms over the same dataset indicates that Random Forest and Decision Tree have the highest accuracy in prediction of heart disease based on the dataset that we have provided. Similarly, Naive Bayes has the least accurate results for this scenario under the given contexts.

Throughout the past twenty years, we've seen a huge increase in the number of school universities. Given the intense competition among major universities and schools, this attracts students to apply for admission to these institutions. Early school dropout prediction is a critical problem for learners, and it is hard to tackle. And a wide number of factors can impact student retention. In order to attain the best accuracy, the conclusion of the program, the standard classification approach that was used to solve this problem frequently needs to be applied the majority of organizations and courses launched by universities operate on either an auto model, therefore they always prefer course enrollment over student caliber. As a result, many students stop taking the course after the first year. In order to manage student dropout rates, this research provides a data mining application. The predictive model may provide an effective predictive list of students who typically require the greatest help from the student dropout program given updated data on new students. The results indicate that the object classification algorithm Random Forest data mining technique can create a reliable prediction model using existing student academic data. Future research on student dropout rates will continue to be vital for informing policy decisions, identifying at-risk populations, evaluating interventions, enhancing support services, predicting trends, understanding long-term consequences, and promoting global learning and collaboration in education.

Predicting crop yields is one of the more difficult tasks in the agriculture sector. A fascinating area of research to estimate agricultural productivity has emerged from recent advancements in information technology for agriculture. Crop yield prediction is a technique for estimating crop production based on a variety of factors, including weather conditions and parameters such as temperature, rainfall, fertilizer, and pesticide use. In the world of agriculture, Data mining techniques are extremely popular. In order to predict the crop production for the following year, data mining techniques are employed and evaluated in the agricultural sector. In this paper, we carried out the comparison between Naive Bayes, K-nearest neighbor, Decision Tree, Random Forest, and K-Means clustering algorithms to predict crop yield in order to determine which method is most effective at doing so. The results show which algorithm is better suitable for this particular purpose by comparing these data mining algorithms for agricultural crop production and determining which algorithm is more successful for this outcome.

Almost all educational institutions have shifted their academic activities to digital platforms due to the recent COVID-19 epidemic. Because of this, it is very important to assess how well teachers are performing with this new way of online teaching. Educational Data Mining (EDM) is a new field that emerged from using data mining techniques to analyze educational data and making decision based on findings. EDM can be utilized to gain better understanding about students and their learning processes, assist teachers do their academic tasks, and make judgments about how to manage educational system. The primary objective of this study is to uncover the key factors that influence the quality of teaching in a virtual classroom environment. Data is gathered from the students’ evaluation of teaching from computer science students of three online semesters at X University. In total, 27622 students participated in these survey. Weka, sentimental analysis, and word cloud generator are applied in the process of carrying out the research. The decision tree classifies the factors affecting the performance of the teachers, and we find that student-faculty relation is the most prominent factor for improving the teaching quality. The sentimental analysis reveals that around 78% of opinions are positive and “good” is the most frequently used word in the opinions. If the education system is moved online in the future, this research will help figure out what needs to be changed to improve teachers’ overall performance and the quality of their teaching.

Cardiovascular disease is the leading cause of death. In recent days, most people are living with cardiovascular disease because of their unhealthy lifestyle and the most alarming issue is the majority of them do not get any symptoms in the early stage. This is why this disease is becoming more deadly. However, medical science has a large amount of data regarding cardiovascular disease, so this data can be used to apply data mining techniques to predict cardiovascular disease at the early stage to reduce its deadly effect. Here, five data mining classification techniques, such as: Naïve Bayes, K-Nearest Neighbors, Support Vector Machine, Random Forest and Decision Tree were implemented in the WEKA tool to get the best accuracy rate and a dataset of 12 attributes with more than 300 instances was used to apply all the data mining techniques to get the best accuracy rate. After doing this research people who are at the early stage of cardiovascular disease or probably going to be a victim can be identified more accurately.

Scaled agile approaches are increasingly being used by automotive businesses to cope with the complexity of their organizations and products. The development of automotive systems necessitates the use of safe procedures. SafeScrum® is a real example of how agile approaches may be used in the creation of high-reliability systems on a small scale. A framework like SAFe or LeSS does not facilitate the creation of safety-critical systems in large-scale contexts from the start. User stories are a wonderful approach to convey flexible demands, the lifecycle is iterative, and testing is the initial stage in the development process. Scrum plus extra XP approaches may be used to build high-reliability software and certification by the IEC 61508 standard is required for the software. This adds a slew of new needs to the workflow. Scrum's quality assurance measures proved to be inadequate in a recent industry situation. Our study's overarching goal is to provide light on the Scrum development process so that it may be improved for use with life-or-death systems. Our study of the business world was a mixed-methods affair. The findings demonstrated that although Scrum is helpful in ensuring the security of each release, it is less nimble in other respects. The difficulties of prioritization, communication, time constraints, and preparing for and accepting new safety standards were all discussed. In addition, we have had some helpful feedback from the business world, but the generality issue arising from this particular setting has yet to be addressed.

After release of Web 2.0 in 2004 user spawned contents on the internet eminently in abundant review sites, online forums, online blogs, and many other sites. Entire user generated contents are considerable bunches of unorganized text written in different languages that encompass user emotions about one or more entities. Mainly predictive analysis exerts the existing data to forecast future outcomes. Currently, a massive amount of researches are being engrossed in the area of opinion mining, also called sentiment analysis, opinion extraction, review analysis, subjective analysis, emotion analysis, and mood extraction. It can be an utmost choice whilst perceiving the meaning and patterns in prevailing data. Most of the time, there are various algorithms available to work with polling. There are contradictory opinions among researchers regarding the effectiveness of algorithms. We have compared different opinion mining algorithms and presented the findings in this paper.

During the recent Covid-19 pandemic, there has been a tremendous increase in online-based learning (e-learning) activities as nearly every educational institution has transferred its programs to digital platforms. This makes it crucial to investigate student performance under this new mode of delivery. This research conducts a comparison among the traditional educational data mining techniques to detect the best performing classifier for analyzing as well as predicting students’ performance in online learning platforms during the pandemic. It is achieved through extracting four datasets from X-University student information system and learning platform, followed by the application of 6 classifiers to the extracted datasets. Random Forest Classifier has demonstrated the highest accuracy in the first two out of the four datasets, while Simple Cart and Naïve Bayes Classifiers presented the same for the remainder two. All the classifiers have demonstrated medium to high TP rates, class precision and recall, ranging from 60% to 100% for almost all of the classes. This study emphasized the attributes that have a direct impact on students’ performance. The outcomes of this study will assist the instructors and educational institutions to identify important factors in the analysis and prediction of student performance for online program delivery.

COVID-19 hit the world unexpectedly, forcing humans to isolate themselves. It has placed the lives of people in jeopardy with its fury. The global pandemic had a detrimental effect on the worlds' education spheres. It has imposed a global lockdown, with a negative impact on the students' lives. Continuing regular classes on-campus was out of the question. At that moment, online learning came to us as a savior. The quality of online education was yet to be tested on a large scale compared to regular schooling. Educational data mining is a modern arena that holds promise for those who work in education. Data mining strategies are developed to uncover latent information and identify valuable trends that can increase students' performance and, in turn, contribute to the improvement of the educational system in the long run. This research mainly aims to identify a comparative analysis of the students' academic performance between online and on-campus environments and distinguish the significant characteristics that influence their academic endeavors. The impact of the factors on the students' performance is visualized with the help of the Decision Tree Classification Model. This paper will assist in giving a good overview that influences the distinguished factors on students' academic performance. Moreover, educators will also be benefited from this paper while making any important decision regarding the educational activity.

In modern communication scenario of the 5G era, the service quality is the greatest concern for the users. Also, the concept of security can’t be neglected in this case. In the IoT oriented services like vehicle and VANET systems, the security in the presentation layer of the network is required. This work is over the security mechanism of the service storage and fetching the files for service. A new scheme of multi layered file and content encryption has been produced in order to strengthen the security of the file and data to maintain integrity and confidentiality of the IoT enabled services implemented in 5G. The encryption scheme is designed for the password encryption through asymmetric key cryptography (RSA) along with an enhanced concern of internal content or data security with symmetric key (AES-128) cryptography. This encryption system of double layer for a file makes the study unique and differentiable than other security schemes.

Network intrusion is an increasing major concern as we are rapidly advancing in technology. To detect network intrusion, Intrusion Detection Systems are required. Among the wide range of intrusion detection technologies, machine learning methods are the most appropriate. In this paper we investigated different machine learning techniques using NSL-KDD dataset, with steps of building a model. We used Decision Tree, Support Vector Machine, Random Forest, Naïve Bayes, Neural network, adaBoost machine leaning algorithms. At step one, one-hot-encoding is applied to convert categorical to numeric features. At step two, different feature scaling techniques, including normalization and standardization, are applied on these six selected machine learning algorithms with the encoded dataset. Further in this step, for each of the six machine learning algorithms, the better scaling technique application outcome is selected for the comparison in the next step. We considered six pairs of better scaling technique with each machine learning algorithm. Among these six scaling-machine learning pairs, one pair (Naïve Bayes) is dropped for having inferior performance. Hence, the outcome of this step is five scaling-machine learning pairs. At step three, different feature reduction techniques, including low variance filter, high correlation filter, Random Forest, Incremental PCA, are applied to the five scaling-machine learning pairs from step two. Further in this step, for each of the five scaling-machine learning pairs, the better feature reduction technique application outcome is selected for the comparison in the next step. The outcome of this step is five feature reduced scaling-machine learning pairs. At step four, different sampling techniques, including SMOTE, Borderline-SMOTE, ADASYN are applied to the five feature reduced scaling-machine learning pairs. The outcome of this step is five over sampled, feature reduced scaling-machine learning pairs. This outcome is then finally compared to find the best pairs to be used for intrusion detection system.

Among the many music information retrieval (MIR) tasks, music genre classification is noteworthy. The categorization of music into different groups that came to existence through a complex interplay of cultures, musicians, and various market forces to characterize similarities between compositions and organize collections is known as a music genre. The past researchers extracted various hand-crafted features and developed classifiers based on them. But the major drawback of this approach was the requirement of field expertise. However, in recent times researchers, because of the remarkable classification accuracy of deep learning models, have used similar models for MIR tasks. Convolutional Neural Net- work (CNN), Recurrent Neural Network (RNN), and the hybrid model, Convolutional - Recurrent Neural Network (CRNN), are such prominently used deep learning models for music genre classification along with other MIR tasks and various architectures of these models have achieved state-of-the-art results. In this study, we review and discuss three such architectures of deep learning models, already used for music genre classification of music tracks of length of 29-30 seconds. In particular, we analyze improved CNN, RNN, and CRNN architectures named Bottom-up Broadcast Neural Network (BBNN) [1], Independent Recurrent Neural Network (IndRNN) [2] and CRNN in Time and Frequency dimensions (CRNN- TF) [3] respectively, almost all of the architectures achieved the highest classification accuracy among the variants of their base deep learning model. Hence, this study holds a comparative analysis of the three most impressive architectural variants of the main deep learning models that are prominently used to classify music genre and presents the three architecture, hence the models (CNN, RNN, and CRNN) in one study. We also propose two ways that can improve the performances of the RNN (IndRNN) and CRNN (CRNN-TF) architectures.

The paper projected to study the field of m-learning focusing on investigating the facilities required to initiate an m-learning environment. Facilities and regular practices of conventional learning and e-learning was considered to find the potential facilities for m-learning environment. We used Integrated Tertiary Educational Supply Chain Model framework that stands on conventional education and illustrates the combined form of education supply chain and research supply chain model. Two surveys were conducted to collect data from students and teachers of higher education. The responses from both of the surveys have been presented and later compared with the findings from our studies of the existing learning environments. The significance of this research is in identifying the facilities for a learner and educator centric m-learning environment.

The Air Pressure System (APS) is a type of function used in heavy vehicles to assist braking and gear changing. The APS failure dataset consists of the daily operational sensor data from failed Scania trucks. The dataset is crucial to the manufacturer as it allows to isolate components which caused the failure. However, missing values and imbalanced class problems are the two most challenging limitations of this dataset to predict the cause of the failure. The prediction results can be affected by the way of handling these missing values and imbalanced class problem. In this paper, we have examined and presented the impact of five different missing value imputation techniques namely: Expectation Maximization, Mean Imputation, Soft Impute, MICE, and Iterative SVD in producing significantly better results. We have also performed an empirical comparison of their performance by applying five different classifiers namely: Naive Bayes, KNN, SVM, Random Forest, and Gradient Boosted Tree on this highly imbalanced dataset. The primary aim of this study is to observe the impact of the mentioned missing value imputation techniques in the enhancement of the prediction results, performing an empirical comparison to figure out the best classification model and imputation technique. We found that the MICE imputation and the random under-sampling techniques are the highest influential techniques for improving the prediction performance and false negative rate.

Every software project is unique in its own way. As a consequence, a single software process model cannot be suitable for all types of projects. In the real world, practitioners face different difficulties with the existing process models during development. Still, they cope up with the challenges by tailoring the software development lifecycle according to their needs. Most of these custom-tailored practices are kept inside the walls of the organizations. However, sharing these proven and tested practices as well as acquired knowledge and experience would be highly beneficial for other practitioners as well as researchers. So in this paper, we have presented a software process model which contains the characteristics of both Scrum and Waterfall model and named it “ScrumFall”. This model has been practicing in an Anonymous Software Development Company, Bangladesh to solve the shortcomings of Scrum and Waterfall models. Moreover, we have analyzed the performance and suitability for applying this process model. The result shows that this process model is highly effective for the certain projects.

Automated Vehicular System has become a necessity in the current technological revolution. Real Tra?c sign detection and recognition is a vital part of that system that will ?nd roadside tra?c signs to warn the automated system or driver beforehand of the physical conditions of roads. Mostly, researchers based on Tra?c sign detection face problems such as locating the sign, classifying it and distinguishing one sign from another. The most common approach for locating and detecting tra?c signs is the color information extraction method. The accuracy of color information extraction is dependent upon the selection of a proper color space and its capability to be robust enough to provide color analysis data. Techniques ranging from template matching to critical Machine Learning algorithms are used in the recognition process. The main purpose of this research is to give a review based on methods and framework of Traffic Sign Detection and Recognition solution and discuss also the current challenges of the whole solution.

In this paper, we have tried to identify the prominent factors of Rice production of all the three seasons of the year (Aus, Aman, and Boro) by applying K-Means clustering on climate and soil variables' data warehoused using Fact Constellation schema. For the clustering, the popular machine-learning tool Weka was used whose visualization feature was principally useful to determine the patterns, dependencies, and relationships of rice yield on different climate and soil factors of rice production.

Software testers should prioritize test cases so that important ones are run earlier in the regression testing process to reduce the cost of regression testing. Test case prioritization techniques schedule test cases for execution in an order that improves the performance of regression testing. One of the performance goals i.e. the fault detection rate, measures how quickly faults are detected during the testing process. Improved rate of fault dependency detection can provide faster feedback on software and let developers debug the leading faults at first that cause other faults to appear later. Another performance goal i.e. severity detection rate among faults, measures how quickly more severe faults are detected earlier during testing process. Previous studies addressed the second goal, but did not consider dependency among faults. In this paper an algorithm is proposed to prioritize test cases based on rate of severity detection associated with dependent faults. The aim is to detect more severe leading faults earlier with least amount of execution time and to identify the effectiveness of prioritized test case.

Advances in computer mediated communication technologies have sparked and continue to facilitate the proliferation of online courses, degree programs, and educational institutions. Leading the way with these advances has been the use of asynchronous discussion forums. However merely setting up a discussion forum does not always ensure quality participation and interaction. The way the course is managed has an impact on the participation as well. This paper compares the difference in course management over four study periods and discusses the resulting consequences on the participation and achievement of the students. This paper also investigates the quality of interaction as perceived by fully online students. The main benefits of this research are that it provides a guideline regarding what course management factors can make the difference in online participation in fully online courses, and how the quality of interaction can be designed.

The rapid advancement of fully online courses has made education accessible from “anywhere” and “anytime”. One of the major success factors of online courses is effective student - content interaction - which defines how students interact with the content in the fully online learning environment. However, there appears to have been little research published about the relationship of content to course design and consequent outcomes for students. In this article, we report on our research on the investigation of factors that affect student - content interaction in fully online computing courses. We have conducted our research through surveys and used a grounded theoretic approach for data analysis. Our results identify the factors that are perceived by students as critical in ensuring effective student - content interaction in fully online computing courses.

K-means algorithm is one of the most popular algorithms for data clustering. With this algorithm, data of similar types are tried to be clustered together from a large data set with brute force strategy which is done by repeated calculations. As a result, the computational complexity of this algorithm is very high. Several researches have been carried out to minimize this complexity. This paper presents the result of our research, which proposes a modified version of k-means algorithm with an improved technique to divide the data set into specific numbers of clusters with the help of several check point values. It requires less computation and has enhanced accuracy than the traditional k-means algorithm as well as some modified variant of the traditional k-Means.