Browse Items (16481 total)
Sort by:
-
A Dimensionality Reduction Model: A Retrospective Approach on Dementia Triggering Parameters and Feature Ranking
The medical sector has advanced in an imposing way, and are coming up with lifesaving models and wearable devices for disease predictions and patient monitoring. The prediction models and wearable devices will lead to immense amount of data collection leading to the dimensionality issues, overfitting and inaccurate results. From the pool of data that we use for our prediction model, we should be able to identify the required information and parameters which gives a positive contribution to the decision making model. Every dataset with higher number of parameters and high dimensionality will tend to the problems of overfitting. Here, we have a dataset of demented and non-demented patients with five conventional features and other physical parameters. Along with these parameters, we are adding three new prediction parameters like glyhb, BMI and Cholesterol, for proving the association of Diabetics and Dementia. After the addition of these parameters, the dataset will have thirty parameters, and dimensionality reduction is done to avoid the condition of overfitting. The work uses Principal Component Analysis(PCA)for reducing the dimensionality, t-SNE for visualization and K means clustering is used to cluster the target variable. The cluster mean of each variable is used to understand the performance of each variable in each cluster. Later, a basic feature ranking method is also implemented which can be further used for the prediction model. The performance metric used in this research work is Silhouette score, Inertia and Inter-Cluster Distance map. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Predicting Stock Market Trends: Machine Learning Approaches of a Possible Uptrend or Downtrend
This paper delves into a statistical analysis of the stock market, emphasizing the significance of accuracy in stock predictions. Large data sets can be handled by machine learning algorithms, which can also forecast outcomes based on past data and spot intricate patterns in financial data. They assist control risks, automate decision-making procedures, and adjust to changing circumstances. Multi-source data can be combined by ML models to provide a comprehensive picture of market circumstances. They can manage intricate, nonlinear interactions, provide impartial analysis, and lessen human bias. Models are able to adjust to shifting market conditions through ongoing learning and retraining. They must, however, exercise caution when deploying models in real-world situations and ensure that they are validated. Although machine learning has advantages for stock market analysis, it must be carefully evaluated for dangers and validated before being used in practical situations. The traditional machine learning model, Logistic Regression has been used in order to predict stock prices. It focuses on binary classification based on the trend of the stock. Through the model training and evaluation and additional analysis done on the results, this research contributes towards obtaining predictions and studying reasons of a possible uptrend or downtrend to further assist companies. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Complex Network Articulation Points Detection and Centrality Measures
To clearly understand how network structure and function interact is a basic difficulty in the study of large networked systems. An old-fashioned idea from graph theory, called articulation points, may be used to do this. In a network, a node If removing it causes the network to become disconnected or causes more network components to get linked, it is an articulation point (AP). Single points of collapse are represented as articulation points in networks. The major goal of this research is to provide a method for identifying the articulation points and centrality measures. We can locate the articulation points considerably more quickly and effectively by using TARJANS Algorithm, which uses depth-first search. It must fulfill two requirements to qualify as an articulation point. For the root node of a DFS traversal to be an articulation point, it must contain at least two offspring nodes that are members of various sub graphs. It has been discovered that articulation points (APS) are crucial for maintaining the reliability and connection of several real-world networks. By assigning each node in the graph a scalar value based on an assumption, centrality metrics may be used to quantify each nodes significance. A fundamental centrality metric is node degree. In terms of node neighbors, it is equivalent. Hence, the more neighbors a node has, the more central and densely linked it is, and the more it affects the network by having more neighbors. ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2025. -
Quantitative Structure-Activity Relationship Modeling for the Prediction of Fish Toxicity Lethal Concentration on Fathead Minnow
As there has been a rise in the usage of in silico approaches, for assessing the risks of harmful chemicals upon animals, more researchers focus on the utilization of Quantitative Structure Activity Relationship models. A number of machine learning algorithms link molecular descriptors that can infer chemical structural properties associated with their corresponding biological activity. Efficient and comprehensive computational methods which can process huge set of heterogeneous chemical datasets are in demand. In this context, this study establishes the usage of various machine learning algorithms in predicting the acute aquatic toxicity of diverse chemicals on Fathead Minnow (Pimephales promelas). Sample drive approach is employed on the train set for binning the data so that they can be located in a domain space having more similar chemicals, instead of using the dataset that covers a wide range of chemicals at the entirety. Here, bin wise best learning model and subset of features that are minimally required for the classification are found for further ease. Several regression methods are employed to find the estimation of toxicity LC50 value by adopting several statistical measures and hence bin wise strategies are determined. Through experimentation, it is evident that the proposed model surpasses the other existing models by providing an R2 of 0.8473 with RMSE 0.3035 which is comparable. Hence, the proposed model is competent for estimating the toxicity in new and unseen chemical. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Assessing Player Interaction for a Social Networking Cooperative Educational Game
Cooperative interaction in educational games, designed to stimulate teamwork, joint creativity and knowledge sharing, also carries potential security threats. One of the key dangers is data leakage. Player interaction involves the exchange of information, and in case of insufficient protection of the system, confidential data, such as personal information, game progress results or individual strategies, may become available to unauthorized persons. This may result in misuse of information, damage to reputation and violation of player privacy. The impact on the game space is also a threat. By interacting, players can change the game world, for example, by entering incorrect data, moving objects to an inappropriate location, or modifying the rules of the game. This can lead to a violation of the balance of the game, incorrect results and a deterioration in the learning effect. Substitution or falsification of game elements is no less dangerous. Attackers can introduce fake elements into the game space, for example, incorrect reviews, changed rules or incorrect data. This can lead to incorrect conclusions, distort learning outcomes, and undermine confidence in the game. In addition, the use of interaction tools can become an object of attack. Attackers can hack and modify tools, such as communication platforms or data storage systems. This can lead to data theft, incorrect operation of tools and malfunction during the game. It is shown that formal descriptions of the choice of a game strategy can exist in a game. Indicators that are essential for cooperative interaction are determined, and examples of their calculation for the case with remote interaction through a social network are given. The article contains information about collaborations, which can be used to assess and choose the direction of development in projects that use game cooperative strategies to implement tasks other than training. The project highlights aspects of cooperative interaction that affect the formation of game strategies in an educational project. Of particular interest are projects in which a social network is the tool and medium of interaction. The objectives of the project are to identify easy-to-use indicators that show the features of cooperative interaction within an educational game. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Flipped Classroom Strategy in Online Teaching: Challenges Faced by Higher Education Teachers
Flipped classroom model has gained increasing interest among university teachers in recent years [1] (Stohr et al.). The reason for its popularity is attributed to its bearing on Vygotskys constructivism theory and for the student centered approach [2] (Ziling Xu et al.). Countries in the world are affected by COVID-19 including India. Hence higher education institutes have begun their online classes. Flipped classroom teaching has been quite prevalent in Indian higher education recently. Online class initiation from higher education institutes in India has pushed faculty members to teach online and faculty have begun flipped classroom teaching online. Flipped classroom teaching in online differs from the face to face mode. There are challenges and issues while using flipped classroom in online mode by the faculty members of higher education. This leads to the present study to find out the challenges of flipped classroom teaching in online mode by teachers of higher education. The present study adopted qualitative research method. Structured interviews and focus group discussion were conducted to answer the research question. Study was able to discuss the challenges of flipped classroom in online mode. These challenges are to be dealt with by the stake holders to bring teaching efficacy. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Pre-Service and In-Service Teachers Perceptions of Using Virtual Reality Tools in Teaching
This paper explores pre-service and in-service teachers perceptions of virtual reality (VR) technology as a teaching and learning tool in the classroom in India. The study aimed to answer four research questions, including the adoption rate of VR technology among teachers, their confidence levels in teaching using VR technologies compared to digital technologies, attitudes towards using VR technology, and the usefulness of different uses of VR technology. The survey conducted among 102 teachers found limited adoption of VR technology, lower confidence levels in using it, but willingness to use it in the future. The paper recommends providing adequate training and support to increase teachers confidence in using VR technology in their teaching practices. The study also suggests that strategies to promote VR technology should consider gender differences in attitudes towards it. Overall, the research concludes that teachers view VR technology as having potential benefits for learning and teaching across various uses. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Gamification and Game-Based Learning: A Systematic Review and Comparative Analysis
In the modern world, characterized by the rapid development of technology and digitalization of almost all spheres of life, it is necessary to keep up with the times and gradually introduce information technology into our lives. This will allow us to remain competitive in a changing world, take advantage of new opportunities and improve our quality of life. It is important to understand that information technology is not just a fashion trend, but a necessary tool for successful development and progress. The paper examines the very concept of gamification, the main methods of introducing gamification into education, highlights the advantages of learning with the addition of gamification, and also works on comparing learning with and without gamification elements. The introduction of game elements into the educational process helps to improve the perception of educational material, as well as increase the level of motivation of the students themselves. It is worth noting that the learning process with the addition of game elements helps to improve attention, develop logical thinking, as well as analyze various situations. Gamification can be viewed from several angles. For a teacher, this teaching method will help to capture the attention of children, which will help create a working atmosphere in the classroom. And for students, gamification is a great opportunity to explore really important topics in game mode. They will have an increased interest in learning, which will have a beneficial effect on their further academic performance and learning. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Cognitive Engagement Scale (CES) in an Online Environment: Construction and Validation
Researchers have demonstrated linkages between active engagement of students with learning material and greater learning gains. Cognitive engagement is a significant component of educational experience. Understanding the challenges associated with cognitive engagement and measuring cognitive engagement in a MOOC environment is challenging. It is the need of the hour with online learning being equivalent to classroom learning in todays dynamic academic environment. The present study aims to construct cognitive engagement scale (CES) to measure the cognitive engagement of learners who sign up for the massive open online courses (MOOC). The aim of this study is dual-fold: firstly, to conceptualize the cognitive dimension of learner engagement within MOOCs, and secondly, to construct a theoretically informed scale for assessing cognitive engagement in online environments. Study presents a detailed process of the scale development, which included item generation, item evaluation, pilot testing, testing psychometric properties of the scale, and scale refinement. The researchers crafted the initial questionnaire drawing from both existing literature and personal insights. Subject matter experts then validated the items within the questionnaire and ensured its reliability through a pilot study, where it was administered to a sample of 100 participants The final version of the scale captures the four dimensions of cognitive engagement: Passive receiving, active manipulating, constructive generating, and interactive dialoguing. The present study contributes to the growing literature on cognitive engagement and adds to the existing literature of MOOC engagement scale with focus on cognitive engagement exclusively. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Unbalanced Dataset Preprocessing Using Hybrid Combination Algorithm for Arrhythmia Detection
Nowadays timely vaticination of cardiovascular conditions with the aid of a computer-backed opinion system minimizes the mortality. Cardiac arrhythmia discovery is one of the most grueling tasks because the variations of electrocardiogram (ECG) signal are veritably slight, which can not be recognized by mortal eyes. The data set under disquisition in this work is taken from the well-known Arrhythmia Dataset, which is codified into different classes. The correct identification of the health condition can lead to an easier, more effective, and less precious reclamation. This proffered study incorporates times of exploration on arrhythmia discovery exercising coincidental technologies. This project addresses the challenge of class imbalance in the analysis of arrhythmia data, utilizing hybrid combination of thr?? approaches like Tomek Links, RUSBoost Classifier, and ADASYN. Here, used the hybrid combination of these three methods.The study commences with data preprocessing, including the loading and preparation of the dataset. Feature extraction and label separation are performed to enable further analysis. Subsequently, resolve th? dataset into groups of testing and training to facilitate robust model valuation. Additionally, this article provides a thorough analysis of new preprocessing model approaches for diagnosing heart disease. By comparing the performance of th?s? methods, this research contributes to th? development of robust and accurate arrhythmia classification models, with potential applications in clinical diagnostics and healthcare decision support systems. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Optimizing Healthcare Analytics: A Zero-Inflated Poisson Approach toPediatric Emergency Room Visits
In various fields, the modeling of count data holds significant importance. The Poisson regression model is a commonly utilized tool for this purpose. However, this model assumes that the data has uniform dispersion, a condition often not met in real-world observations. The nature of overdispersion canvary depending on the specific context. When the overdispersion is primarily dueto an excessive number of zero counts, the Zero-inflated Poisson regressionmodel becomes a more suitable choice for modeling count data. The paper commencesby offering a summary of the theoretical foundations of both Poisson regressionand Zero-inflated Poisson regression. To evaluate their performance, usethe Mean-Squared error (MSE) as a comparative metric. Next, apply these modelsto analyze the frequency of hospital emergency room visits by children between 1018 years of age. The overdispersion of the visit count in our dataset is mostly caused by the excessive occurrence of zero counts. The findings demonstratethat the Zero-inflated Poisson regression model outperforms the standard Poisson regression model in terms of MSE. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Enhancing Education Policy Estimation: A Novel Ridge Fuzzy Regression Approach for Handling Multicollinearity with Fuzzy Input Data
Multicollinearity often complicates regression analysis, both in classical and fuzzy input setup. This research introduces a new approach that combines ridge regression with fuzzy regression to tackle correlated covariates impact, with a specific focus on improving education policy systems. Our method utilizes the ?-level estimation algorithm and a dataset where Grade Point Average (GPA) serves as a fuzzy input, while input variables remain crisp. We assess our estimators performance using RMSE and MAPE. This applied research showcases the potential of our method in enhancing education policies through more accurate data-driven decision-making. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Emerging Era of Artificial Intelligence Based Chatbots: Is It Really Required to Today Generation?
The advent of AI-powered chatbots has brought about a significant wave of technological advancement, influencing different aspects of our everyday existence. This study rigorously investigates the necessity of AI-powered chatbots in the present-day generation. Although chatbots provide the possibility for increased convenience, efficiency, and accessibility, their extensive use prompts inquiries on their necessity, ethical issues, and the possible impact on human interaction and creativity. This study attempts to provide a thorough analysis of the benefits and limitations of AI chatbots, highlighting their importance and relevance in the present period. It emphasizes the necessity of a balanced approach towards integrating these digital companions. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Gender and Ethnicity Recognition System Based on Convolutional Neural Networks
The classification of Gender and Ethnicity has been utilized in diverse scenarios, specifically in the realm of human-computer interaction, visual surveillance, and electronic customer services. Predicting the gender and ethnicity of individuals presents a significant obstacle due to its complex characteristics. The escalating prevalence of social media has emphasized the utmost importance of independently predicting gender and race. In this research endeavor, a framework is utilized which utilizes a Convolutional Neural Network to forecast gender and ethnicity by utilizing various outputs starting from the initial stage. The models performance was evaluated using different metrics, including the F1-score, accuracy, precision, recall, and accuracy. The methodology is evaluated using the UTKFace dataset for predicting gender and ethnicity, and compared the model with previous study to understand which model is giving better accuracy. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Advancing Healthcare Decision Support: Leveraging Fuzzy DEMATEL for Delivering Quality Care
Healthcare Decision Support Systems (DSS) play a pivotal contribution in modern healthcare, aiding in informed decision-making and the distribution of high-quality care. To optimize the systems, it is critical to recognize and prioritize the enablers that provide to their successful establishment and operation. This study presents a comprehensive analysis of 10 key enablers essential for the development and deployment of healthcare DSS for quality care. Utilizing Fuzzy DEMATEL (Decision-Making Trial and Evaluation Laboratory), a powerful methodology for discovering complex interdependencies among factors, we systematically evaluate the relationships among these enablers. The enablers, ranging from data integration and clinical collaboration to privacy safeguards and continuous improvement mechanisms, are scrutinized through the lens of Fuzzy DEMATEL, which accommodates the inherent uncertainties and ambiguities within healthcare data. The findings from the study shed light on the strength and direction of the relationships among the enablers, unveiling critical factors that exert substantial influence and those that are most susceptible to external changes. By applying Fuzzy DEMATEL, this study backs to a deeper understanding of the multifaceted nature of healthcare DSS development, offering insights to guide decision-makers, healthcare practitioners, and system developers in their pursuit of improved DSS that enhance the quality of healthcare delivery. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
IOT Wearable Medical Device for Heart Disease Recognition Based ML and DL: A Classification Approach
In the past few years, heart disease has become the foremost worldwide contributor to mortality. This ailment, with a profound effect on the functioning of the heart, leads to issues such as infections in the coronary arteries and diminished blood vessel performance. These complications can culminate in severe unlikely events like heart attacks and strokes. In India alone, approximately one person succumbs to heart disease every minute. To curb the fatalities stemming from cardiac disorders, there is an urgent need for a swift and efficient detection strategy. IoT sensors are utilized in conjunction with Machine Learning (ML) and Deep Learning (DL) techniques to identify heart disease. In this research, we have successfully applied IoT devices and a sensor network to detect heart diseases. This study introduces a medical IoT device designed to gather heart data from patients both before and after the onset of heart disease. This continuously transmitted data is processed using RBF, MLP, and Bi-LSTM models for predicting heart disease. The deep learning approach utilizes past analyses to learn critical features related to heart disease, achieving efficiency in handling complex data. After conducting a series of experiments, we evaluate the systems performance using metrics such as f-measure, sensitivity, specificity, loss function, and Receiver Operating Characteristic (ROC) curves. The HDRBi-LSTM method, in combination with IoT-based analysis, achieves an impressive accuracy rate of 99.5% with minimal time complexity (5 s), effectively reducing heart disease mortality by simplifying the diagnosis of this condition. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Enhanced Approach for Precision Agriculture Using AI/ML Techniques
Precision-based agriculture has been made possible by recent technical breakthroughs and developments in information technology. These new developments have made it possible to better utilise contemporary methods and instruments, like IOT, soft computing, and wireless sensor technology, to increase the agricultural productions environmental and economic sustainability. Precision farming is a new trend in agriculture that sets itself apart from traditional farming methods by applying resources in a way that is efficient, planned, systematic, and justified in order to produce higher and better yields. Precision farming uses geographic information systems like weather patterns, remote sensing technologies like Wireless Sensor Networks (WSN), and soft computing tools like Support Vector Machines (SVM), Random Forest (RF), Artificial Neural Networks (ANN), and Decision Trees (DT) to monitor and predict farm produce requirements in real time and for the future. This study examines the application of several methods and tools used in precision farming. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Unveiling Dynamics of Structural Breaks in Global Stock Markets and Implications for Forecasting Accuracy
This research investigates structural breaks in global stock markets, employing the Chow test on major indices from January 2013 to November 2023. Results reveal significant breaks in NYSE (November 2020) linked to the US election and positive vaccine trials, Nasdaq (May 2020) amidst global concerns over COVID-19, and Euronext 100 (February 2021), suggesting market shifts. Notably, Shanghai Stock Exchange experienced a robust break in December 2014, contrasting with SZSE's non-significant break. HKEX experiences a significant shift in June 2020, possibly influenced by US regulatory policies and COVID-19. The Nifty index shows a profound break in December 2020, correlated with pandemic severity. LSE Group evidences a break in July 2019, while the Saudi Exchange shows non-significant evidence in March 2021. The study underscores the importance of considering structural breaks for accurate market forecasting and decision-making. Descriptive statistics provide insights into market characteristics. The methodology integrates the Chow test and CUSUM squares for break detection. Findings contribute to understanding global market dynamics and emphasize the impact of external events on structural stability. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Enhancing User Control: A Reinforcement Learning Framework for Breaking Filter Bubbles in Recommender Systems
In an age of information overload, recommendation systems play an important role in providing personalized content to users. However, traditional recommendation systems often create filter bubbles, limiting the types of content users are exposed to. Based on the research presented in the article Breaking the Filter Bubble: A Reinforcement Learning Framework for Controllable Recommender Systems, this article proposes a new approach to further improve the controllability and diversity of recommendations. By using reinforcement learning techniques, the proposed framework aims to break the filter bubble by providing users with more diverse content recommendations while maintaining high recommendation accuracy. Extensive experiments on real-world datasets demonstrate the effectiveness of this approach in suppressing recommendation concentration and improving recommendation diversity. The results of this study contribute to the further development of controllable recommendation systems and provide insights into solving the filter bubble problem in recommendation systems. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Secure Communication in Fog Nodes through Quantum Key Distribution
In this intensifying era of big-data, data processing is predominant. Data processing takes place closer to devices in the fog computing, strong security measures are necessary. The Hybrid Encapsulation Mechanism (HEM), Elliptic Curve Diffie-Hellman (ECDH), and Quantum Key Distribution (QKD) are furnished in this study as a safe method for authentic data sharing between fog nodes. HEM, a special encapsulation technique that makes use of the Advanced Encryption Standard (AES), is used to encrypt the data, and ECDH is used to generate the shared and private keys. A conventional route is then used to convey the data, which may offer a compromise between efficiency and security. But, fog nodes can communicate and exchange shared keys more easily over a quantum channel, which is an efficient quantum key exchange mechanism (QKD). ECC efficiently and rapidly generates encryption keys to safeguard data and these keys are impenetrable by even the most powerful computers due to their proven security. Along with safeguarding private information, this promotes collaboration and trust across fog nodes. The efficiency of the proposed method is calculated considering the execution time, computation cost, key size, and key strength. We discovered that our proposed approach outperforms existing ones when compared to them. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.