Search

Abstract The educational landscape is witnessing a transformation with the integration of Educational Technology (Edutech). As educational institutions adopt digital platforms and tools, the generation of Educational Big Data (EBD) has significantly increased. Research indicates that educational institutions produce massive data, including student enrollment records, academic performance metrics, attendance records, learning activities, and interactions within digital learning environments. This influx of data needs efficient processing to derive actionable insights and enhance the learning experience. Real-time data processing has a critical part in educational environments to support various functions such as personalized learning, adaptive assessment, and administrative decision-making. However, there may be challenges in sending large amounts of educational data to cloud servers, i.e., latency, cost and network congestion. These challenges make it more difficult to provide educators and students with timely insights and services, which reduces the efficiency of educational activities. This paper proposes a Regional Computing (RC) paradigm designed specifically for big data management in education to address these issues. In this case, RC is established within educational regions and intended to decentralize data processing. To reduce dependency on cloud infrastructure, these regional servers are strategically located to collect, process, and store big data related to education regionally. Our investigation results show that RC significantly reduces latency to 203.11 ms for 2,000 devices, compared to 707.1 ms in Cloud Computing (CC). It is also more cost-efficient, with a total cost of just 1.14 USD versus 5.36 USD in the cloud. Furthermore, it avoids the 600% congestion surges seen in cloud setups and maintains consistent throughput under high workloads, establishing RC as the optimal solution for managing EBD.

Abstract The healthcare sector is experiencing a digital transformation propelled by the Internet of Medical Things (IOMT), real-time patient monitoring, robotic surgery, Electronic Health Records (EHR), medical imaging, and wearable technologies. This proliferation of digital tools generates vast quantities of healthcare data. Efficient and timely analysis of this data is critical for enhancing patient outcomes and optimizing care delivery. Real-time processing of Healthcare Big Data (HBD) offers significant potential for improved diagnostics, continuous monitoring, and effective surgical interventions. However, conventional cloud-based processing systems face challenges due to the sheer volume and time-sensitive nature of this data. The migration of large datasets to centralized cloud infrastructures often results in latency, which impedes real-time applications. Furthermore, network congestion exacerbates these challenges, delaying access to vital insights necessary for informed decision-making. Such limitations hinder healthcare professionals from fully leveraging the capabilities of emerging technologies and big data analytics. To mitigate these issues, this paper proposes a Regional Computing (RC) paradigm for the management of HBD. The RC framework establishes strategically positioned regional servers capable of regionally collecting, processing, and storing medical data, thereby reducing dependence on centralized cloud resources, especially during peak usage periods. This innovative approach effectively addresses the constraints of traditional cloud processing, facilitating real-time data analysis at the regional level. Ultimately, it empowers healthcare providers with the timely information required to deliver data-driven, personalized care and optimize treatment strategies.

This paper conducts a thorough comparative analysis of optimization algorithms for an unconstrained convex optimization problem. It contrasts traditional methods like Gradient Descent (GD) and Nesterov Accelerated Gradient (NAG) with modern techniques such as Adaptive Moment Estimation (Adam), Long Short-Term Memory (LSTM) and Multilayer Perceptron (MLP). Through empirical experiments, convergence speed, solution accuracy and robustness, is evaluated providing insights to aid algorithm selection. The convergence dynamics of convex optimization, is explored analyzing classical algorithms and contemporary neural network (NN) methodologies. The study concludes with a comparative assessment of these algorithms performance metrics and their respective strengths and weaknesses.

In recent years, rapid advancements in deepfakes (incorporating Artificial Intelligence (AI), machine, and deep learning) have updated tools and techniques for manipulating multimedia. Though technology has primarily been utilized for beneficial purposes, such as education and entertainment, it is also used for malicious or unethical tasks to spread disinformation or ruin someone’s dignity, even if it encompasses harassing and blackmailing victims. Deepfakes refer to high-quality and realistic multimedia-manipulated content that has been digitally modified or synthetically generated. We conducted a systematic literature review of deepfake detection to offer an updated overview of existing research work that initially describes the widely accessible deepfake generation tools, classifications, and detection process. We highlighted recent techniques in visual deepfake detection based on the feature representations, grouped into four domains: spatial, temporal, frequency, and spatio-temporal, including their key features and limitations by providing details of existing datasets, together with the potentials of deepfake and its future directions. This study tried to add an updated repository of technological change in deepfake, which could help researchers to develop robust deepfake models.

The Internet of Things (IoT) has connected millions of devices to the internet for communication and computation purposes. Due to constraints such as limited battery power, processing capabilities, and storage capacity, IoT devices often rely on remote fog nodes for task execution, a practice known as task offloading. The NP-hard nature of offloading and scheduling IoT tasks onto fog nodes poses significant challenges. Current IoT task offloading often relies on single parameters like device MIPS, RAM, or battery power. Similarly, task allocation on fog nodes frequently prioritizes VM MIPS. To address energy consumption and task failures in IoT systems, robust fault tolerance, efficient task offloading, and effective scheduling are crucial. This paper introduces a novel Fault-Tolerant, Priority-Based Task Offloading and Scheduling Model (FP-TOSM) for IoT logistics. Utilizing the Analytic Hierarchy Process (AHP) for multi-criteria decision-making, IoT tasks are prioritized. Tasks below a specified threshold are executed locally, while those within a defined range are offloaded to a fog node. Tasks exceeding this range are delegated to the cloud. The Euclidean formula is used to determine the proximity between the IoT logistics vehicle and fog nodes for offloading decisions. Dynamic re-clustering fault tolerance mechanisms are utilized to manage task failures after offloading. Successful offloaded tasks are allocated to suitable VMs in the cloud and fog nodes. The proposed strategy selects the most suitable VM for offloaded tasks using a multi-criteria decision-making process to reduce energy consumption, SLA violations, and execution costs. The model's performance is evaluated through simulations in iFogSim2. Results demonstrate reductions in energy consumption by up to 5.8 % and 10.4 %, decreases in SLA violations by up to 25.2 %, and a 16.28 % improvement in response time compared to an ACO-based model with a response time of 17.8 %. Additionally, the task failure ratio shows a significant reduction of up to 22.1 %. These findings highlight the effectiveness of FP-TOSM in enhancing efficiency and reliability within fog computing environments.

Abstract The latest advancements of deep learning have resulted in a new era of natural language processing. The machines now possess an unparallel ability to interpret and engage with various tasks such as text classification, content generation and natural language understanding. This development extended to the analysis of human behavior, where deep learning models are used to decode human personality. Due to the rise of social media, generating huge amounts of textual data that reshaped communication patterns. Understanding personality traits is a challenging topic which helps us to explore the patterns of thoughts, feelings and behaviors which are helpful for recruitment, career counselling and consumers’ behavior for marketing, etc. In this research study, the main aim is to predict the human personality trait of agreeableness showing whether a person is emotional who feels a lot or thinker who is logical and has rational thinking. This behavior leads to analyzing them as cooperative, friendly and respecting difference of views. For comprehensive empirical analysis, shallow machine learning models, ensemble models, and deep learning technique including state of the art transformer-based models are applied on real-world dataset of MBTI. For feature engineering, textual features of TF-IDF and POS tagging and word embeddings such as word2vec, glove and sentence embeddings are explored. The results analysis shows the highest performance 91.57% with sentence embeddings utilizing Bi-LSTM algorithm that highlights the power of this study as compared to existing studies in the relevant literature.

Abstract Background Anomaly detection is crucial in healthcare data due to challenges associated with the integration of smart technologies and healthcare. Anomaly in electronic health record can be associated with an insider trying to access and manipulate the data. This article focuses around the anomalies under different contexts. Methodology This research has proposed methodology to secure Electronic Health Records (EHRs) within a complex environment. We have employed a systematic approach encompassing data preprocessing, labeling, modeling, and evaluation. Anomalies are not labelled thus a mechanism is required that predicts them with greater accuracy and less false positive results. This research utilized unsupervised machine learning algorithms that includes Isolation Forest and Local Outlier Factor clustering algorithms. By calculating anomaly scores and validating clustering through metrics like the Silhouette Score and Dunn Score, we enhanced the capacity to secure sensitive healthcare data evolving digital threats. Three variations of Isolation Forest (IForest)models (SVM, Decision Tree, and Random Forest) and three variations of Local Outlier Factor (LOF) models (SVM, Decision Tree, and Random Forest) are evaluated based on accuracy, sensitivity, specificity, and F1 Score. Results Isolation Forest SVM achieves the highest accuracy of 99.21%, high sensitivity (99.75%) and specificity (99.32%), and a commendable F1 Score of 98.72%. The Isolation Forest Decision Tree also performs well with an accuracy of 98.92% and an F1 Score of 99.35%. However, the Isolation Forest Random Forest exhibits lower specificity (72.84%) than the other models. Conclusion The experimental results reveal that Isolation Forest SVM emerges as the top performer showcasing the effectiveness of these models in anomaly detection tasks. The proposed methodology utilizing isolation forest and SVM produced better results by detecting anomalies with less false positives in this specific EHR of a hospital in North England. Furthermore the proposal is also able to identify new contextual anomalies that were not identified in the baseline methodology.

In this paper, a quick convergence adaptive structure is proposed for trajectory tracking of an articulated type robotic manipulator using the Hardware in the Loop (HIL) simulation technique. A novel Nonlinear Quick Convergence Adaptive Sliding Backstepping Control (QCASBC) algorithm is implemented on a C2000 real-time controller board. The performance of the proposed control algorithm is inspected concerning the sliding mode PID (SM-PID) control technique to estimate its correlation with the proposed algorithm. The experimental part of the HIL simulation tests has been carried out on a simulated model of a three-link serial robot manipulator. A dynamic model of the robotic manipulator has been developed using Matlab Simulink software and its performance is analyzed using the HIL technique via a C2000 real-time controller for tracking the desired trajectory. Results show that the speed of convergence while tracking the desired trajectory of the manipulator is better in the proposed algorithm. It is also estimated that the positional error of the QCASBC algorithm is superior to the SM-PID control algorithm. The observed average angle error is improved by 14% and the average response time error is improved by 11% by using the proposed QCASBC algorithm compared to the SM-PID approach.

Abstract Genes, expressed as sequences of nucleotides, are susceptible to mutations, some of which can lead to cancer. Machine learning and deep learning methods have emerged as vital tools in identifying mutations associated with cancer. Thyroid cancer ranks as the 5th most prevalent cancer in the USA, with thousands diagnosed annually. This paper presents an ensemble learning model leveraging deep learning techniques such as Long Short-Term Memory (LSTM), Gated Recurrent Units (GRUs), and Bi-directional LSTM (Bi-LSTM) to detect thyroid cancer mutations early. The model is trained on a dataset sourced from asia.ensembl.org and IntOGen.org, consisting of 633 samples with 969 mutations across 41 genes, collected from individuals of various demographics. Feature extraction encompasses techniques including Hahn moments, central moments, raw moments, and various matrix-based methods. Evaluation employs three testing methods: self-consistency test (SCT), independent set test (IST), and 10-fold cross-validation test (10-FCVT). The proposed ensemble learning model demonstrates promising performance, achieving 96% accuracy in the independent set test (IST). Statistical measures such as training accuracy, testing accuracy, recall, sensitivity, specificity, Mathew's Correlation Coefficient (MCC), loss, training accuracy, F1 Score, and Cohen's kappa are utilized for comprehensive evaluation.

Hate speech constitutes a major problem on microblogging platforms, with automatic detection being a growing research area. Most existing works focus on analyzing the content of social media posts. Our study shifts focus to predicting which users are likely to become targets of hate speech. This paper proposes a novel Hate-speech Target Prediction Framework (HTPK) and introduces a new Hate Speech Target Dataset (HSTD), which contains tweets labeled for targets and non-targets of hate speech. Using a combination of Term Frequency-Inverse Document Frequency (TFIDF), N-grams, and Part-of-Speech (PoS) tags, we tested various machine learning algorithms, Naïve Bayes (NB) classifier performs best with an accuracy of 93%, significantly outperforming other algorithms. This research identifies the optimal combination of features for predicting hate speech targets and compares various machine learning algorithms, providing a foundation for more proactive hate speech mitigation on social media platforms.

Abstract Background Healthcare programs and insurance initiatives play a crucial role in ensuring that people have access to medical care. There are many benefits of healthcare insurance programs but fraud in healthcare continues to be a significant challenge in the insurance industry. Healthcare insurance fraud detection faces challenges from evolving and sophisticated fraud schemes that adapt to detection methods. Analyzing extensive healthcare data is hindered by complexity, data quality issues, and the need for real-time detection, while privacy concerns and false positives pose additional hurdles. The lack of standardization in coding and limited resources further complicate efforts to address fraudulent activities effectively. Methodolgy In this study, a fraud detection methodology is presented that utilizes association rule mining augmented with unsupervised learning techniques to detect healthcare insurance fraud. Dataset from the Centres for Medicare and Medicaid Services (CMS) 2008-2010 DE-SynPUF is used for analysis. The proposed methodology works in two stages. First, association rule mining is used to extract frequent rules from the transactions based on patient, service and service provider features. Second, the extracted rules are passed to unsupervised classifiers, such as IF, CBLOF, ECOD, and OCSVM, to identify fraudulent activity. Results Descriptive analysis shows patterns and trends in the data revealing interesting relationship among diagnosis codes, procedure codes and the physicians. The baseline anomaly detection algorithms generated results in 902.24 seconds. Another experiment retrieved frequent rules using association rule mining with apriori algorithm combined with unsupervised techniques in 868.18 seconds. The silhouette scoring method calculated the efficacy of four different anomaly detection techniques showing CBLOF with highest score of 0.114 followed by isolation forest with the score of 0.103. The ECOD and OCSVM techniques have lower scores of 0.063 and 0.060, respectively. Conclusion The proposed methodology enhances healthcare insurance fraud detection by using association rule mining for pattern discovery and unsupervised classifiers for effective anomaly detection.

Cybersecurity, complimenting authentication, has become the backbone of the Internet of Things. In the authentication process, the word authentication is of the utmost importance, as it is the door through which both Mr. Right Guy and Mr. Wrong Guy can pass. It is the key to opening the most important and secure accounts worldwide. When authentication is complete, surely there will be passwords. Passwords are a brain-confusing option for the user to choose when making an account during the registration/sign-up process. Providing reliable, effective, and privacy-preserving authentication for individuals in mobile networks is challenging due to user mobility, many attack vectors, and resource-constrained devices. This review paper explores the transformation and modern mobile authentication schemes, categorizing them into password, graphical, behavioral, keystroke, biometric, touchscreen, color, and gaze-based methodologies. It aims to examine the strengths and limitations focused on challenges like security and usability. Standard datasets and performance evaluation measures are also discussed. Finally, research gaps and future directions in this essential and emerging area of research are discussed.

Social networks have enabled the exchange of information that has accelerated the diffusion of online content. Social networks offer multiple forms of social interaction that facilitate users in generating online content. The increasing popularity of social networks has attracted many researchers to investigate their impact on users and confirmed significant influence of social networks on markets, social life, and politics. Therefore, identification of the influential users in social networks has become a popular research area due to several of its applications in diverse domains such as e-commerce, viral marketing, political campaigns etc. However, earlier research either consider linguistic content or utilize network-based representation for finding the influential users in social networks. This paper incorporates association rule mining (ARM) based algorithms, that are mostly used for market-basket analysis, for exploring the behavior of users and predict their participation in social interactions. The study identifies topic-based influential users using Apriori and FPGrowth on the dataset of a popular online question-answer community. In addition, the study employs standard evaluation metrics such as confidence, support, lift, and conviction for computing association rules. Subsequently, the obtained results are validated using conventional measures including accuracy, precision, and recall in accordance with the perspective of association rules. The experiments are performed on multiple social networking datasets and the obtained results prove the visibility and quality of the proposed method against well-established degree centrality, PageRank, and HITS. In addition, the results validate the effectiveness of ARM in identifying influential users in social networks. The generated rule set can help create efficient decision support systems that are becoming prevalent in providing solutions to real-world problems.

The emergence of online enterprises spread across continents has given rise to the need for managing the tacit knowledge and expertise of employees. Scenarios that includes the intention of the employer to find tacit expertise and knowledge of an employee that is not documented or self-disclosed have been addressed in this article. In today’s world management of tacit knowledge has become important for the organizations. Recent studies have also proposed hosting tacit knowledge management module over cloud for large global enterprises. There are many conceptual frameworks for the acquisition and processing of the tacit knowledge. This article has proposed a reputation based approach utilizing social interactions of employees to identify the expert based on the tacit knowledge. The existing reputation-based approaches towards expertise ranking in enterprises utilize PageRank, Normal distribution, and the Hidden Markov model for expertise ranking. These models suffer negative referral, collusion, reputation inflation, and dynamism. However, the authors have proposed a Bayesian-based approach utilizing the beta probability distribution reputation model for employee ranking in enterprises that can be hosted as a cloud service for the employees of the enterprise. The experimental results reveal improved performance compared to previous techniques in terms of mean average error (MAE) for the three data sets. The proposed scheme is able to differentiate categories of interactions in a dynamic context. The results reveal that the technique is independent of the rating pattern and density of data.

This research addresses the imperative need for advanced detection mechanisms for the identification of phishing websites. For this purpose, we explore state-of-the-art machine learning, ensemble learning, and deep learning algorithms. Cybersecurity is essential for protecting data and networks from threats. Detecting phishing websites helps prevent fraud and safeguard personal information. To evaluate the efficacy of our proposed method, the top features using information gain, gain ratio, and PCA are used to predict and identify a website as phishing or non-phishing. The proposed system is trained using a dataset that covers 11,055 websites. The ensemble learning model applied achieved an impressive 99% accuracy in predicting phishing websites, surpassing previous models, and setting a new benchmark in the field. The findings highlight the effectiveness of combining deep learning architectures with ensemble learning, offering not only improved accuracy but also adaptability to emerging phishing techniques.

Recognizing fraudulent credit card transactions is one of the main issues facing banking institutions. Since each transaction that completes the authentication procedure must be authorized by financial institutions, a hacker might pose as the actual cardholder and execute a fraudulent transaction. In this paper, we investigated the capacity of ensemble learning methods to identify credit card frauds on two distinct data sets: the Sparkov synthetic dataset and the real dataset of consumers in the European Union. XGBoost models, random forests, and naive Bayes classifiers are applied and assessed on both datasets. Accuracy, precision, recall, and F1 score are used to measure performance. According to the results, most ensemble classifiers perform exceptionally well on the real-world dataset, but significantly poorly on the simulated dataset. This study showed that, unlike in simulated environments, credit card transaction management scripts are quickly learned in deterministic settings. It is discussed that a larger danger of card information leakage results from strict determinism and lack of randomness.

The rise of smart health technologies has offered benefits such as remote patient monitoring, personalized therapy, and early disease detection. However this advancement has also raised concerns about security and privacy of patient data. Authentication is essential for protecting patient data and smart health devices. The existing scheme is identified with security vulnerabilities in terms of stolen verifier, impersonation, and man in middle attacks. Motivated by this, authors have presented a secure, lightweight authentication technique for smart health environments for a remote monitoring scenario employing a Multiple Factor approach and a lightweight algorithm. The scheme is verified through formal and informal analysis. The scheme demonstrated enhanced authentication performance and security with processing overhead of (0.04 milliseconds) and transmission overhead of (704 bits), in comparison to the baseline scheme. This system has the potential to improve data security and privacy in a variety of smart health scenarios.

In 2022, the digital population reached 5.4 billion, leading to network congestion, primarily attributed to the widespread use of social media, robotic programs, and prime-time television during peak hours. This congestion posed a significant challenge, as sensitive services struggled to operate effectively during peak hours, while network resources remained largely underutilized during off-peak hours. To tackle this issue, we propose the implementation of Regional Servers. These servers are designed to initially store less critical data during peak hours and subsequently transfer it to the cloud during off-peak hours, effectively optimizing network resources. The results demonstrate that the deployment of regional servers proved highly efficient in mitigating delays and reducing jitters that users typically experience during peak hours, successfully addressing the challenges posed by network congestion.

Ensuring the safety of vehicles requires the critical responsibility of diagnosing and correcting brake faults. Implementing this proactive measure to address brake faults not only ensures the protection of lives but also enhances the efficiency and cost-effectiveness of repair processes conducted on-site. Machine learning technology has recently contributed to a significant rise in the popularity of predictive maintenance. The objective of this study is to provide a method for identifying issues with the air pressure system (APS) of air brake systems in heavy-duty vehicles. The data obtained by sensors has been used to analyse the APS failure in this Scania Truck. After examining numerous classification methods, Random Forest was determined to have the greatest performance, with a classification accuracy of 99.4%. Moreover, the implementation of eXplainable Artificial Intelligence has included the use of SHapley Additive exPlanation (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) to provide explanations for the contributions of features in model predictions. We picked 20 features from the wheel speed sensor data received from several Internet of Things (IoTs) sensors, which significantly influenced our final selection. By repeatedly applying random forest to these 20 features, we achieved the same degree of accuracy as previously. Consequently, our suggested approach used a reduced amount of computer resources and was less intricate to execute in terms of calculation.

Climate change’s impact on human health poses unprecedented and diverse challenges. Unless proactive measures based on solid evidence are implemented, these threats will likely escalate and continue to endanger human well-being. The escalating advancements in information and communication technologies have facilitated the widespread availability and utilization of social media platforms. Individuals utilize platforms such as Twitter and Facebook to express their opinions, thoughts, and critiques on diverse subjects, encompassing the pressing issue of climate change. The proliferation of climate change-related content on social media necessitates comprehensive analysis to glean meaningful insights. This paper employs natural language processing (NLP) techniques to analyze climate change discourse and quantify the sentiment of climate change-related tweets. We collected a total number of 5506 tweets for the period of January 2022 and February 2023 and manually labeled them to make the dataset for this experiment. ClimateBERT, a pre-trained model fine-tuned specifically on the climate change domain was used to generate the context vectors. Several machine learning algorithms with different feature encoding techniques, such as TF-IDF and BERT, have been implemented to classify user sentiments. When comparing the performance of the classifiers using different evaluation metrics such as precision, recall, accuracy, and f-measure, the ClimateBERT + Random Forest model is found to be outperforming all the other baselines with an accuracy of 90.22%, recall of 85.22%, and an f-measure of 85.47%. The findings from this experiment unearth valuable insights into public sentiment and the entities associated with climate change discourse. Policymakers, researchers, and organizations can leverage such analyses to understand public perceptions, identify influential actors, and devise informed strategies to address climate change challenges.

Addressing the challenge of missing values is a critical step when preparing and analyzing data. This process, known as imputation, helps ensure the dataset is complete, accurate, and reliable. As a result, the possibility of bias and errors in subsequent analysis is significantly reduced. The key contribution of this work is to assess the efficiency of imputation by feature importance employing several base learning algorithms. This study investigates the effectiveness of individual and ensemble machine learning methods as the base learning algorithms, including support vector machines with the linear kernel (SVML), boosted linear regression (BLR), deep boost (DBP), and K-Nearest Neighbor (K-NN), in predicting missingness patterns. The dataset for each category explicitly introduces missingness patterns, including missing not at random (MNAR), at random (MAR), and completely at random (MCAR) at different percentages (15%, 45%, 25%), and Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE) are among the commonly used performance matrices employed to gauge the effectiveness of the IBFI framework. The dataset for this study, comprising soil radon and thoron gas concentration time series along with meteorological parameters, the dataset spans a 14-month period. Four earthquake events were recorded during the whole study period. The deep boosting model (DBP) consistently outperforms other base learning models in imputing missing values across various variables within the imputation by feature importance (IBFI) framework. Specifically, DBP achieves an average RMSE value of 573.165 for the Radon variable under MCAR scenarios. For the Thoron variable, DBP demonstrates impressive performance with average MAPE values of 0.7405, 0.7249, and 0.8212 under MCAR, MNAR, and MAR conditions respectively. Additionally, DBP yields competitive results for imputing missing entries in Temperature, Relative Humidity, and Pressure variables. These findings highlight effectiveness of DBP in accurately predicting missing values. This study concludes that the IBFI with deep boosting model executes the imputations quite accurately relative to other base learning models. Moreover, this study recommends using DBP as a base learning algorithm in imputation by feature importance framework for uncovering hidden patterns in time series data like soil radon gas. The replication of the study using heterogeneous datasets would enhance the understanding of the generalization and broader applicability of the imputation by feature importance.

Smart devices in smart cities face the dual challenge of requiring real-time data processing and storing that data permanently on the cloud for future use. This creates a significant conflict between the need for immediate responsiveness and the demands of long-term storage. While Cloud Computing (CC) offers a viable platform for processing and storing this data, it introduces delays, longer response times, and network congestion issues. Internet of Things (IoT) devices require real-time responses, but their data must also be available for subsequent analysis and decision-making. Although edge and fog computing paradigms have been developed to address these issues, they suffer from limited resources and scalability challenges, making them inadequate for the above needs. A hybrid paradigm is often proposed to address the limitations of both cloud and edge computing. However, synchronization between cloud and edge can introduce delays, negatively affecting overall performance. The proposed framework, Regional Computing (RC), addresses big data challenges in smart cities by providing an intermediate solution between edge (limited power) and cloud (distant) computing. RC manages data during peak hours, alleviating congestion on the public network. The collected data is subsequently offloaded to cloud servers for long-term storage and detailed analysis. Preliminary findings are very encouraging, indicating substantial enhancements in reducing network congestion, minimizing delays, and lowering costs.

The recent rise of Artificial Intelligence (AI) has already revolutionized human lives and is improving the quality of human life in many ways. The field of AI, Natural Language Processing (NLP), helps to understand, comprehend and even generate new content. NLP is used for various content analysis tasks such as sentiment analysis, fake news detection, etc. However, human personality traits detection is a new research domain. Analyzing users generated content has a significant role in identifying and understanding users’ views and behaviors. Users’ traits detection can be helpful in analysis of consumers’ personalization, finding top candidates for recruitment, career counselling, etc. In this research study, our aim is to predict personality of extroversion behaviors using machine and deep learning approaches. Extroversion means whether a person is an introvert or extrovert as this trait is relevant to certain jobs like team management, social ties etc. For empirical analysis, we investigate MBTI dataset with various feature engineering techniques including textual features like Term Frequency-Inverse Document Frequency (TF-IDF), Parts of Speech (PoS) tagging, as well as deep word embeddings ok word2vec, GloVe. The state-of-the-art shallow machine learning, ensemble modelling and deep learning models are applied. The main novelty is the exploration of latest sentence embeddings which captures semantic information of content in a better manner. Thus, the comprehensive results analysis reveals sentence embeddings as features to Bi-LSTM achieves highest accuracy of 92.52% and outperforms the existing studies in the relevant literature.

This research paper presents the development and analysis of a multifaceted smart grid prototype. It combines various technologies for the smart grid operation. The first technology is environmental analysis of smart grid and solar panel cleaning. Secondly, radio-frequency identification (RFID)-based security and access control system has been integrated for smart grid. The third component is internet of things (IoT)-based energy monitoring and load management. For environmental analysis sensors such as temperature, humidity, light-dependent resistor, and flame sensors are connected to a NodeMCU controller for real time monitoring. Moreover, IoT based solar cleaning system is developed in the form of prototype with the help of Blynk and servo motor. The second component of prototype is smart security system which is developed with the help of Arduino and RFID module to facilitate secure access control. The third part of prototype employs voltage and current sensors with an ESP32 microcontroller and the Blynk application for real-time energy consumption analysis. This setup enables remote monitoring of voltage, power dynamics, and consumption patterns in a smart grid. It also offers an IoT based solution for load management and load shedding within the smart grid. The complete prototype overall demonstrates a comprehensive approach to 1) smart grid management, 2) environmental analysis, 3) security, and 4) energy monitoring.

This paper presents a novel approach for detecting abuse on Twitter. Abusive posts have become a major problem for social media platforms like Twitter. It is important to identify abuse to mitigate its potential harm. Many researchers have proposed methods to detect abuse on Twitter. However, most of the existing approaches for detecting abuse look only at the content of the abusive tweet in isolation and do not consider its contextual information, particularly the tweets posted before the abusive tweet. In this paper, we propose a new method for detecting abuse that uses contextual information from the tweets that precede and follow the abusive tweet. We hypothesize that this contextual information can be used to better understand the intent of the abusive tweet and to identify abuse that content-based methods would otherwise miss. We performed extensive experiments to identify the best combination of features and machine learning algorithms to detect abuse on Twitter. We test eight different machine learning classifiers on content- and context-based features for the experiments. The proposed method is compared with existing abuse detection methods and achieves an absolute improvement of around 7%. The best results are obtained by combining the content and context-based features. The highest accuracy of the proposed method is 86%, whereas the existing methods used for comparison have highest accuracy of 79.2%.

Software development effort estimation (SDEE) is recognized as vital activity for effective project management since under or over estimating can lead to unsuccessful utilization of project resources. Machine learning (ML) algorithms are largely contributing in SDEE domain, particularly ensemble effort estimation (EEE) works well in rectifying bias and subjectivity to solo ML learners. Performance of EEE significantly depends on hyperparameter composition as well as weight assignment mechanism of solo learners. However, in EEE domain, impact of optimization in terms of hyperparameter tunning as well as weight assignment is explored by few researchers. This study aims in improving SDEE performance by incorporating metaheuristic hyperparameter and weight optimization in EEE, which enables accuracy and diversity to the ensemble model. The study proposed Metaheuristic-optimized Multi-dimensional bagging scheme and Weighted Ensemble (MoMdbWE) approach. This is achieved by proposed search space division and hyperparameter optimization method named as Multi-dimensional bagging (Mdb). Metaheuristic algorithm considered for this work is Firefly algorithm (FFA), to get best hyperparameters of three base ML algorithms (Random Forest, Support vector machine and Deep Neural network) since FFA has shown promising results of fitness in terms of MAE. Further enhancement in performance is achieved by incorporating FFA-based weight optimization to construct Metaheuristic-optimized weighted ensemble (MoWE) of individual multi-dimensional bagging schemes. Proposed scheme is implemented on eight frequently utilized effort estimation datasets and results are evaluated by 5 error metrices (MAE, RMSE, MMRE, MdMRE, Pred), standard accuracy and effect size along with Wilcox statistical test. Findings confirmed that the use of FFA optimization for hyperparameter (with search space sub-division) and for ensemble weights, has significantly enhanced performance in comparison with individual base algorithms as well as other homogeneous and heterogenous EEE techniques.

In eradicating tardy behavior and discipline in obeying school rules by students, the role is not only the subject teacher, but the counseling teacher also has a very important role to overcome tardy students. Being late is one of the most commonly seen rule-breaking behaviors in schools. Although being late is not a criminal act against the law like theft and murder, if handling and follow-up efforts are not given, it will have a negative impact on student development, especially in terms of academic achievement. The phenomenon of MTs Al-Jam'iyatul Washliyah Tembung School shows that there are students who are late, the cause of which often occurs due to general factors, family, personal and school factors as a result affecting student learning achievement at school. Qualitative is the method used with a descriptive approach. The BK teacher's effort to resolve late behavior in students is to provide individual counseling guidance services to students. After getting individual counseling guidance services students experience changes in no longer being late for school, their clothes become more rapid and become more quiet.

Career guidance is an effort to provide direction and coaching in determining the future that will be achieved by students. This study aims to describe and analyze career guidance services to increase motivation to continue their education to the tertiary level for students of SMAN 4 Tebing Tinggi. A qualitative approach is used in this study, with an analytical descriptive study method. As for data collection using the method of observation, interviews, and documentation studies. Furthermore, the data were analyzed using data reduction techniques, data presentation, to draw a conclusion. The results of this study found that the majority of class XII students at SMAN 4 Tebing Tinggi had high motivation to continue their studies at various tertiary institutions, although there were some of them who wanted to do business for reasons of family economic limitations. Through the provision of information services based on career guidance conducted by alumni of SMAN 4 Tebing Tinggi, students can get to know the campus world, choose study programs according to their talents and interests, and the positive impact of continuing their studies to a higher education level. Thus, students can consider making career decisions after graduating from high school with a sense of satisfaction, full awareness, and responsibility for career choices.

Abstract Living organisms including fishes, microbes, and animals can live in extremely cold weather. To stay alive in cold environments, these species generate antifreeze proteins (AFPs), also referred to as ice-binding proteins. Moreover, AFPs are extensively utilized in many important fields including medical, agricultural, industrial, and biotechnological. Several predictors were constructed to identify AFPs. However, due to the sequence and structural heterogeneity of AFPs, correct identification is still a challenging task. It is highly desirable to develop a more promising predictor. In this research, a novel computational method, named AFP-LXGB has been proposed for prediction of AFPs more precisely. The information is explored by Dipeptide Composition (DPC), Grouped Amino Acid Composition (GAAC), Position Specific Scoring Matrix-Segmentation-Autocorrelation Transformation (Sg-PSSM-ACT), and Pseudo Position Specific Scoring Matrix Tri-Slicing (PseTS-PSSM). Keeping the benefits of ensemble learning, these feature sets are concatenated into different combinations. The best feature set is selected by Extremely Randomized Tree-Recursive Feature Elimination (ERT-RFE). The models are trained by Light eXtreme Gradient Boosting (LXGB), Random Forest (RF), and Extremely Randomized Tree (ERT). Among classifiers, LXGB has obtained the best prediction results. The novel method (AFP-LXGB) improved the accuracies by 3.70% and 4.09% than the best methods. These results verified that AFP-LXGB can predict AFPs more accurately and can participate in a significant role in medical, agricultural, industrial, and biotechnological fields.

Penelitian ini bertujuan untuk membuktikan bahwa layanan informasi mampu mengubah stigma negatif terhadap guru BK. Pendekatan kualitatif merupakan metode yang digunakan dalam penelitian ini. Penelitian ini menggunakan data primer dan data sekunder sebagai sumber data, hasil observasi dan wawancara merupakan data primer. Hasilnya terdapat perubahan stigma negatif siswa terhadap guru BK setelah diberikan layanan informasi secara bertahap, dari 33 siswa di MTSN1 Labuhanbatu khususnya dikelas VIII C sebelumnya terdapat 22 siswa yang memiliki stigma negatif terhadap guru BK dan setelah diberikan layanan secara bertahap 10 orang siswa memiliki perubahan stigma negatif secara keseluruhan terhadap guru BK. Faktor pendukung perubahan stigma negatif guru BK tidak terlepas dari dukungan kepala sekolah kepada guru BK dan bersamaan dengan perubahan sikap guru BK menjadi lebih ramah dan mudah tersenyum kepada siswa –siswi di MTSN1 Labuhanbatu, lebih mendekatkan diri kepada siswa –siswi dan menjalin komunikasi yang baik, sementara faktor penghambatnya ialah guru BK tidak mempunyai jam khusus untuk melakukan layanan, kurangnya pemahaman guru mata pelajaran mengenai pentingnya peran dan tugas guru BK

Information retrieval from huge social web data is a challenging task for conventional search engines. Recently, information filtering recommender systems may help to find movies, however, their services are limited because of not considering movie quality aspects in detail. Prediction of movies can be improved by using the characteristics of social web content about a movie such as social-quality, tag quality, and a temporal aspect. In this paper, we have proposed to utilize several features of social quality, user reputation and temporal features to predict popular or highly rated movies. Moreover, enhanced optimization-based voting classifier is proposed to improve the performance on proposed features. Voting classifier uses the knowledge of all the candidate classifiers but ignores the performance of the model on different classes. In the proposed model, weight is assigned to each model based on its performance for each class. For the optimal selection of weights for the candidate classifiers, Genetic Algorithm is used and the proposed model is called Genetic Algorithm Voting (GA-V) classifier. After labeling the suggested features by using a fixed threshold, several classifiers like Bayesian logistic regression, Naïve Bayes, BayesNet, Random Forest, SVM, Decision Tree, LSTM and AdaboostM1 are trained on MovieLens dataset to find high-quality/popular movies in different categories. All the traditional ML models are compared with GA-V in terms of precision, recall and F1 score. The results show the significance of the proposed features and proposed GA-V classifier.

AbstractNamed entity recognition (NER) continues to be an important task in natural language processing because it is featured as a subtask and/or subproblem in information extraction and machine translation. In Urdu language processing, it is a very difficult task. This paper proposes various deep recurrent neural network (DRNN) learning models with word embedding. Experimental results demonstrate that they improve upon current state‐of‐the‐art NER approaches for Urdu. The DRRN models evaluated include forward and bidirectional extensions of the long short‐term memory and back propagation through time approaches. The proposed models consider both language‐dependent features, such as part‐of‐speech tags, and language‐independent features, such as the “context windows” of words. The effectiveness of the DRNN models with word embedding for NER in Urdu is demonstrated using three datasets. The results reveal that the proposed approach significantly outperforms previous conditional random field and artificial neural network approaches. The best f‐measure values achieved on the three benchmark datasets using the proposed deep learning approaches are 81.1%, 79.94%, and 63.21%, respectively.

Deep learning (DL) has attracted increasing attention on account of its significant processing power in tasks, such as speech, image, or text processing. In order to the exponential development and widespread availability of digital social media (SM), analyzing these data using traditional tools and technologies is tough or even intractable. DL is found as an appropriate solution to this problem. In this paper, we keenly discuss the practiced DL architectures by presenting a taxonomy-oriented summary, following the major efforts made toward the SM analytics (SMA). Nevertheless, instead of the technical description, this paper emphasis on describing the SMA-oriented problems with the DL-based solutions. To this end, we also highlight the DL research challenges (such as scalability, heterogeneity, and multimodality) and future trends.

In Urdu, part of speech (POS) tagging is a challenging task as it is both inflectionally and derivationally rich morphological language. Verbs are generally conceived a highly inflected object in Urdu comparatively to nouns. POS tagging is used as a preliminary linguistic text analysis in diverse natural language processing domains such as speech processing, information extraction, machine translation, and others. It is a task that first identifies appropriate syntactic categories for each word in running text and second assigns the predicted syntactic tag to all concerned words. The current work is the extension of our previous work. Previously, we presented conditional random field (CRF)-based POS tagger with both language dependent and independent feature set. However, in the current study, we offer: 1) the implementation of both machine and deep learning models for Urdu POS tagging task with well-balanced language-independent feature set and 2) to highlight diverse challenges which cause Urdu POS task a challenging one. In this research, we demonstrated the effectiveness of machine learning and deep learning models for Urdu POS task. Empirically, we have evaluated the performance of all models on two benchmark datasets. The core models evaluated in this study are CRF, support vector machine (SVM), two variants of the deep recurrent neural network (DRNN), and a variant of n-gram Markov model the bigram hidden Markov model (HMM). The two variants of DRRN models evaluated include forward long short-term memory (LSTM)-RNN and LSTM-RNN with CRF output.

Web content credibility implies finding credible and correct information on the web. Recent studies have shown there is an increasing trend of users turning towards the web for searching information related to a variety of topics including health, stocks, education, politics to name few. Information credibility is a critical factor in these domains for the decision makers. There is no limitation on the authorship of those articles and content. One criterion for evaluating credibility is to check the authority or source of information. However, there are situations when wrong information flows from credible sources. There are various approaches towards credibility assessment, broadly categorized into human-based and computational approaches. Computational approaches utilizing machine learning based techniques are computationally expensive. Reputation based approaches overcome this, however the latest work fails to take into account issue of negative referrals and utilizes simple summation as the calculation structure making it more resilient to attacks. This paper put forth verified hypothesis of direct relationship of credibility to the expertise of entity. Authors proposed a Bayesian based approach using feedback in the form of interaction among the entities to compute their expertise level, thereby showing improved results in terms of Precision, Correlation and Mean Average Error. The experiments are performed on two different datasets, one of the dataset is developed from a survey as the part of the research study. The results from the two experiments show that the reputation ranks are independent of the pattern of ratings and density of data, unlike previous techniques whose results were limited by these factors. The proposed technique gives 27% and 18% more precise results for the two experiments respectively compared to the baseline. The correlation results are also significant in both experiments for the proposed technique with significant values of 0.39 and 0.87 showing a linear relationship between predicted and original data. The paper also discusses the reputation attacks and proposes counter measures to tackle these attacks through simulation results.

Named entity recognition (NER) is an important task in natural language processing, as it is widely featured as a key information extraction sub-task with numerous application areas. A plethora of attempts was made for NER detection in Western and Asian languages. However, little effort has been made to develop techniques for the Urdu language, which is a prominent South Asian language with hundreds of millions of speakers across the globe. NER in Urdu is considered a hard problem owing to several reasons, including the paucity of large, annotated datasets; an inaccurate tokenizer; and the absence of capitalization in the Urdu language. To this end, this study proposed a conditional-random-field-based technique with both language-dependent and language-independent features, such as part-of-speech tags and context windows of words, respectively. As a second contribution, we developed an Urdu NER dataset (UNER-I) in which a large number of NE types were manually annotated. To evaluate the effectiveness of the proposed approach, as well as the usefulness of the dataset, experiments were performed using the dataset we developed and an existing dataset. The results of the experiments showed that our proposed technique outperformed the baseline technique for both datasets by improving the F1 scores by 1.5% to 3%. Furthermore, the results demonstrated that the enhanced dataset was useful for learning and prediction in a supervised learning approach.

Online social databases are rich sources to retrieve appropriate information that is subsequently analyzed for forthcoming trends prediction. In this paper, we identify rising stars in cricket domain by employing machine learning techniques. More precisely, we predict rising stars from batting as well as from bowling realms. For this intent, the concepts of co-players, team, and opposite teams are incorporated and distinct features along with their mathematical formulations are presented. For classification purpose, generative and discriminative machine learning algorithms are employed, and two models from each category are evaluated. As a proof of applicability, the proposed approach is validated experimentally while analyzing the impact of individual features. Besides, model and categorywise assessment is also performed. Employing cross validation, we demonstrate high accuracy for rising star prediction that is both robust and statistically significant. Finally, ranking lists of top ten rising cricketers based on weighted average, performance evolution, and rising star scores are compared with the international cricket council rankings.

Mining research community on the basis of hidden relationships present between its entities is important from academic recommendation point of view. Previous approaches discovered research community by using network connectivity based distance measures (no text semantics) or by using poorer text semantics and relationships of documents DL (Document Level) by ignoring richer text semantics and relationships of VL (Venue Level). In this paper, we address this problem by considering richer text semantics and relationships. We propose a VAT (Venue Author Topic Approach) based on Author-Topic model to discover inherent community structures in a more realistic way by modeling from VL. We show how topics and authors can be inferred for new venues and how author-to-author and venue-to-venue correlations can be discovered. The positive relationship of topic denseness with ranking performance of proposed approach is explained. Experimental results on research collaborative network \"DBLP\" demonstrate that proposed approach significantly outperformed the baseline approach in discovering community structures and relationships in large-scale network.

Social networking has revolutionized the use of conventional web and has converted World Wide Web into the social web as users can generate their own content. This change has been possible due to social web platforms like forums, wikis, and blogs. Blogs are more commonly being used as a form of virtual communication to express an opinion about an event, product or experience and can reach a large audience. Users can influence others to buy a product, have certain political or social views, etc. Therefore, identifying the most influential bloggers has become very significant as this can help us in the fields of commerce, advertisement and product knowledge searching. Existing approaches consider some basic features, but lack to consider some other features like the importance of the blog on which the post has been created. This paper presents a new metric, MIIB (Metric for Identification of Influential Bloggers), based on various features of bloggers' productivity and popularity. Productivity refers to bloggers' blogging activity and popularity measures bloggers' influence in the blogging community. The novel module of BlogRank depicts the importance of blog sites where bloggers create their posts. The MIIB has been evaluated against the standard model and existing metrics for finding the influential bloggers using dataset from the real-world blogosphere. The obtained results confirm that the MIIB is able to find the most influential bloggers in a more effective manner.

THE GOLDEN ROAD How ancient India transformed the world WILLIAM DALRYMPLE 496pp. Bloomsbury. 30 [pounds sterling]. It is not often that a book on ancient India is aimed at a [...]

History is too important to be left to the (academic) historians. We have been hearing this quite a bit of late. In India's current political climate, deriding historians has become [...]