Browse Items (11809 total)
Sort by:
-
Non-Alcoholic Fatty Liver Disease Prediction with Feature Optimized XGBoost Model
Non-alcoholic fatty liver disease (NAFLD) is an expanding health threat, posing significant risks for long-term complications. Early detection and intervention are crucial, but traditional diagnostic methods can be expensive and invasive.This study investigates the utilization of machine learning models for predicting liver diseases from various out-sourced datasets..We employed Decision Trees, Random Forests, and Support Vector Machines (SVMs) to predict NAFLD based on various clinical and demographic features. Model performance was evaluated by calculating accuracy, precision,deviation and accuracy-score.All these models achieved promising accuracy levels, ranging from 80% to 90%, showcasing their potential for NAFLD prediction. Among them, XG-Boost demonstrated the highest performance, with an accuracy of 90% and more.This study demonstrates the effectiveness of machine learning models in predicting NAFLD with high accuracy using readily available data. Further research with larger sized and more varied datasets will vindicate these models for real-world application in clinical settings. 2024 IEEE. -
Hybrid Deep Learning Based GRU Model for Classifying the Lung Cancer from CT Scan Images
Lung cancer is a potentially fatal condition, posing significant challenges for early detection and treatment within the healthcare domain. Despite extensive efforts, the etiology and cure of cancer remain elusive. However, early detection offers hope for effective treatment. This study explores the application of image processing techniques, including noise reduction, feature extraction, and identification of cancerous regions within the lung, augmented by patient medical history data. Leveraging machine learning and image processing, this research presents a methodology for precise lung cancer categorization and prognosis. While computed tomography (CT) scans are a cornerstone of medical imaging, diagnosing cancer solely through CT scans remains challenging even for seasoned medical professionals. The emergence of computer-assisted diagnostics has revolutionized cancer detection and diagnosis. This study utilizes lung images from the Lung Image Database Consortium (LIDC-IDRI) and evaluates various image preprocessing filters such as median, Gaussian, Wiener, Otsu, and rough body area filters. Subsequently, feature extraction employs the Karhunen-Loeve (KL) methodology, followed by lung tumor classification using a hybrid model comprising a One-Dimensional Convolutional Neural Network (1D-CNN) and a Gated Recurrent Unit (GRU). Experimental findings demonstrate that the proposed model achieves a sensitivity of 99.14%, specificity of 90.00%, F -measure of 95.24%, and accuracy of 95%. 2024 IEEE. -
Data: A Key to HR Analytics for Talent Management
The past few years have witnessed a significant rise in job openings across various industries worldwide. This trend has highlighted the need for companies to plan and recruit better talent to keep up with the demand for skilled employees. As a result, Human Resource (HR) professionals are now using workforce planning and HR analytics to address the challenges of finding and retaining the right employees. With the help of technological advancements in HR systems, businesses are leveraging data and insights to understand workplace dynamics better. Workforce planning has thus become crucial for organizations of all sizes to ensure they have the necessary talent to achieve their goals in the present and future. This chapter delves deeper and examines the importance of workforce planning and how HR analytics can help companies achieve their strategic objectives. Talent Management is about analyzing the workforce skill requirements of the organization. It needs a strategic plan to ensure the appropriate people are in the right roles at the right times. Talent Management is a crucial element of every businesss performance. In this process, data play a pivotal role in evaluating the existing workforce and planning for future workforce needs. Based on this, a strategy is developed to fill gaps or prospective shortages. Organizations can achieve their goals by using talent planning and collecting data about upcoming projects and skill requirements based on market needs. For example, talent planning is essential in the healthcare sector to guarantee that hospitals and clinics have enough doctors, nurses, and other healthcare workers to fulfill the rising demand for healthcare services. HR analytics is the key to talent management, enabling organizations to stay competitive, enhance productivity, and achieve long-term strategic objectives. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Synthesis of Online Criminal User Behaviours Disseminating Bengali Fake News Using Sentiment Analysis
Even though research on artificial intelligence (AI) is still in its early phases, the field is growing in popularity. We created a hybrid machine learning model to better understand the pattern of results connected to illegal user behaviour. Then, after identifying the components of illegal user activity, we created a theory for forecasting criminal user behaviour that explains the patterns and results. Our study focuses on offenders spreading misleading information online and makes use of a Bengali dataset. Sentiment analysis is a modern technology that can help us understand how individuals feel in different scenarios during their everyday lives. To comprehend the pattern behind this agenda, machine learning and deep learning techniques will be applied throughout the categorization process. To determine the possible attitudes driving criminal conduct that spreads misleading information, sentiment levels on social media may be monitored or studied. This study examines the use of several artificial intelligence approaches to assess sentiment in social media data in order to identify criminal user activity occurring throughout the world. The hybrid model CNN with Adam optimizer exhibits higher precision levels while doing sentiment analysis. In addition to identifying solutions to the issues that people currently face in the modern world, we also propose a new categorization system for illicit user activity. In our analysis of the research's shortcomings, we make recommendations for a broader research agenda on illicit user conduct and how one can forecast the criminal user behaviour on psychological aspects. Our model was thus able to draw 87.33% accuracy in determining criminal behaviour patterns. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Cerebral Stroke Classification Using Over Sampling Technique and Machine Learning Models
In recent years, cerebral stroke has ascended as a paramount concern in global public health. Proactive strategies emphasizing metabolic control over salient risk factors present a superior approach compared to relying solely on physiological indicators, which may not delineate clear preventive directives. In this research, we present the SPX-CerebroPredict modela novel machine learning framework designed to classify imbalanced cerebral stroke data for clinical diagnostics. The study delves into feature selection methodologies, employing both information gain and principal component analysis (PCA). To address the class imbalance dilemma, the Synthetic Minority Over-sampling Technique (SMOTE) was harnessed. The empirical evaluation, conducted on the cerebral stroke prediction dataset from Kagglecomprising 43,400 medical records with 783 stroke instancespitted well-established algorithms such as support vector machine, logistic regression, decision tree, random forest, XGBoost, and K-nearest neighbor against one another. The results evince that our SPX-CerebroPredict model, integrating SMOTE, PCA, and XGBoost, surpasses its contemporaries, achieving an impressive accuracy rate of 95%. This discovery underscores the models potential for clinical applicability in cerebral stroke diagnostics. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Unveiling the Future: Exploring Stock Price Prediction in the Finance Sector through Machine Learning and Deep Learning - A Comprehensive Bibliometric Analysis
The investigation of predicting share prices is a captivating and beneficial area of study within the realm of economic research. precise projections and findings can potentially benefit shareholders by reducing the risk of making suboptimal investment selections. The objective of this investigation is to examine the present state of research pertaining to the prognostication of share price predictions through the utilization of Machine Learning (ML) and Deep Learning techniques. The present study examined the existing body of scientific works on methods involving DL and ML in the context of predicting the value of stocks. This study presents a comprehensive overview of research trends, methodologies, and applications in a particular field by conducting a bibliometric analysis of publications indexed in the Scopus database. Drawing from the presented data, recommendations for optimal methodologies can be formulated. The data was visually represented through the utilization of the R programming language and Vos Viewer software. The investigation additionally discerns the primary authors, institutions, and nations that are making contributions to this particular field of research. The outcomes of this investigation possess the potential to guide future research trajectories and offer significant perspectives for professionals and policymakers who are keen on utilizing machine learning and deep learning in the financial sector. 2024 IEEE. -
Emotional Landscape of Social Media: Exploring Sentiment Patterns
Sentiment analysis, a pivotal research area, involves exploring emotions, attitudes, and evaluations prevalent in diverse public spheres. In the contemporary era, individuals extensively share their perspectives on various subjects through social media platforms. Twitter has emerged as a prominent microblogging site, facilitating users to express opinions and insights globally. However, disrespectful or unfair comments have prompted specific platforms to restrict user comments, highlighting the need to foster productive discourse on social media. This study addresses this imperative by analyzing sentiments using data from Twitter. This work employed various deep learning algorithms and methods to classify elements as negative or positive. The Sentiment140 dataset, sourced from Twitter, serves as the training data for the models to identify the most accurate classification approach. By delving into sentiment analysis on Twitter, the study contributes to a better understanding of the nuances of online expressions. It aims to enhance the overall quality of discourse in social media. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Taming theComplexity ofDistributed Lag Models: A Practical Approach toMulticollinearity, Outliers, andAuto-Correlation inFinance
This research investigates the application of robust estimators within the finite distributed lag model (DLM), a critical framework in finance research capturing temporal dependencies between lagged explanatory variables and a response variable. Traditional Ordinary Least Squares (OLS) estimation faces challenges when dealing with high lag counts, multicollinearity, and outliers, potentially compromising parameter estimates and model reliability. Employing real-world data from the RBI, spanning the years 20222023 encompassing budgetary and economic variables of Indian states and Union Territories, the study demonstrates that the MMS estimator emerges as the most efficient estimator, showcasing enhanced robustness against outliers and multicollinearity. Additionally, the study reveals positive autocorrelation in residuals, underscoring the importance of robust methods in addressing such issues in financial modeling. This research contributes valuable insights to financial analysts and offers a more accurate understanding of dynamic relationships in financial systems. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Classification of Diseased Leaves in Plants Using Convolutional Neural Networks
The article focuses on the classification of diseased leaves using a machine learning algorithm. The main focus in agriculture is controlling pests and weeds, for which farmers spray chemical pesticides to get a good yield. The issue here is over-usage and under-usage of pesticides, which might harm the end consumer. To achieve the goal of reducing pesticide use and detecting pests in the crop early, the machine learning algorithm is deployed on the leaf image. The image data of the leaf of the cauliflower plant is collected for 40days. The data was collected from the day the plant was seeded in a pot until the day it was ready to be planted in the soil. From this data, the pest attack on the plants is tracked without the application of pesticides. To achieve this, the CNN algorithm is used on the collected image data. The outcome of the study would be to classify the diseased leaves based on the pest attack and know the right time to spray the pesticides to reduce the damage to the plant. This also reduces the use of pesticides and costs to the farmer. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
A Heuristic Model For Personalised Risk Assesment of PCOS
According to WHO 8-13% of women are affected by Polycystic Ovary Syndrome (PCOS) out of which 70% women remain undiagnosed, it is a common endocrine disorder necessitating early diagnosis for timely intervention. In this paper a heuristic model is developed for PCOS prediction, by combining XGBoost and Random Forest through stacking techniques. Class imbalance was addressed using Random Oversampling. Cross-validation demonstrated the meta-model's superior accuracy compared to individual XGBoost and Random Forest models, highlighting its potential for reliable PCOS prediction. It is observed that the best possible results that the meta-model was able to provide was a score of 93.5% which was acquired in the 4th sample, the lowest score was 87.90% attained in the 2nd sample. To finalise the results, the mean accuracy was calculated which is 90.98% with a standard deviation of 1.96. deterministic model offers reproducible results and interpretability, aiding clinical decision-making. Future research could explore additional biomarkers and probabilistic techniques for personalized risk assessment. 2024 IEEE. -
Comparative Analysis of Machine Learning Models in Predicting Academic Outcomes: Insights and Implications for Educational Data Analytics
In the evolving landscape of educational research, the predictive analysis of student performance using data science has garnered significant interest. This study investigates the influence of diverse factors on academic outcomes, ranging from personal demographics to socioeconomic conditions, to enhance educational strategies and support mechanisms. We employed a diverse ml models to analyze a information containing academic records and socioeconomic information. The models tested include Logistic Regression, Random Forest (RF), Gradient Boosting (GB), Support Vector Machines (SVC), K-Nearest Neighbors (KNN), Gaussian Naive Bayes, and Decision Trees. The process involved comprehensive data preprocessing, exploratory analysis, model training, and evaluation based on metrics such as precision, recall, accuracy, and F1 score. The results indicate that ensemble methods, specifically RF and GB, demonstrate superior efficacy in accurately predicting categories of student performance such as 'Enrolled,' 'Graduated,' and 'Dropped Out.' These models excelled in handling the complex interplay of varied predictors affecting student success. The results further underline the potential of advanced ensemble ML techniques in significantly outperforming the prediction accuracy in the academic domain, hence facilitating the tailoring of educational interventions to foster improved engagement and better outcomes for students. This has provided a comparative analysis of the methods that guide the future application of predictive analytics in education. 2024 IEEE. -
Pixels to Pathogens: A Deep Learning Approach to Plant Pathology Detection
It is known that accurately identifying, early and timely treatment and elimination of the plant diseases is essential for crop protection and healthy crop growth. In traditional or conventional methods, identification and classification were done by testing in laboratories or through visual inspection by farmers. Now going through the testing in labs is very time consuming, while the visual inspection requires enough experience and knowledge. To solve this problem, our study proposes a robust plant pathogen detection method based on a Deep Learning approach on a large dataset containing about 38 categories of different species like Maize, Potatoes, Tomatoes, Bell Pepper, Peach, Strawberry etc. and diseases like rust, molds, blight (late and early). This crop disease detection model leverages the power of the EfficientNetB3 architecture, a state-of-art convolutional neural network(CNN). The main backbone is served by EfficientNetB3and then it is fine-tuned using different hyperparameters and other regularization techniques like weight decay, dropout method and optimizers like RAdam,to enhance the overall accuracy coupled with dynamic learning rate adjustment. In the testing set of the dataset, the proposed model shows encouraging accuracy of about 99.25%, high precision of about 97.35%. A thorough evaluation of the model's functionality is given by the help of training and validation line chart and loss chart that gives the in-depth information on the prediction. And then we implemented the detection model in our mobile application whose interface screen shots are given below. In the application the image can be taken by camera or fed from folders and it will detect the type of disease. 2024 IEEE. -
An Investigation on Machine Learning Models in Classification and Identifications Cervical Cancer Using MRI Scan
This study analyzes the effectiveness of machine learning models in the classification of cervical cancer using a dataset of 900 cancer and 200 non-cancer images gathered from online resources and hospitals. The dataset, covering both CT and MRI images, undergoes rigorous preprocessing, including standardization, normalization, and noise reduction, to enhance its quality for model training. Four machine learning models, namely VGG16, CNN, KNN, and RNN, are recruited to predict cancer and non-cancer cases. During the testing phase, VGG16 emerges as the most accurate, achieving an impressive accuracy of 95.44%, followed by CNN at 92.3%, KNN at 89.99%, and RNN at 86.233%. Performance parameters, such as precision, recall, F1 score, and accuracy, are fully analyzed, providing insights into each model's strengths and capabilities. These discoveries not only contribute to the advancement of cervical cancer diagnostic techniques but also underscore the potential of machine learning in medical imaging. The study emphasizes the relevance of model selection and provides a framework for future research endeavors seeking to enhance the accuracy and performance of cervical cancer diagnosis through the merger of advanced computational techniques with standard diagnostic practices. 2024 IEEE. -
Blockchain Integrated Retail Logistics Chain: An Adoption Perspective
The task of managing public health and safety is a multifaceted and delicate one that demands the careful upkeep of numerous processes and systems, with a particular emphasis on cold chain logistics. The primary objective of this research is to investigate how blockchain technology can meet the needs of a retail cold chain. To accomplish this goal, we employed a comprehensive technological adoption model, partial least squares structural equation modeling, and a quantitative cross-sectional survey approach to ascertain stakeholder adoption intentions toward a blockchain-enabled cold supply chain. Our findings suggest that blockchain technology has the capacity to effectively facilitate the goals of the retail cold chain. 2024 IEEE. -
Sliced Bidirectional Gated Recurrent Unit with Sparrow Search Optimizer for Detecting the Attacks in IoT Environment
In an era characterized by pervasive interconnectivity facilitated by the widespread adoption of Internet of Things (IoT) devices across diverse domains, novel cybersecurity challenges have emerged, underscoring the imperative for robust intrusion detection systems. Conventional security frameworks, constrained by their closed-system architecture, struggle to adapt to the dynamic threat landscape marked by the continual emergence of unprecedented attacks. This paper presents a methodology aimed at mitigating the open set recognition (OSR) challenge within IoT-specific Network Intrusion Detection Schemes (NIDS). Leveraging image-based representations of data, our approach focuses on extracting geographical traffic patterns. We observe that the Recurrent Neural Network exhibits suboptimal classification accuracy and lacks parallelizability for attack analysis tasks. Our investigation concludes that the Sparrow Search Optimization Algorithm (SSOA) serves as a foundation for constructing an effective assault classification model. This research contributes significantly to the field of network security by emphasizing the importance and ramifications of meticulous hyperparameter tuning. It represents a critical stride toward developing IDSs capable of effectively navigating the evolving cyber threat landscape. In the experimental analysis of proposed model reached the accuracy and 0.963% respectively. 2024 IEEE. -
Cost Effective and Energy Efficient Drip Irrigation System for IoT Enabled Smart Agriculture
The conventional methods of smart farming consume a significant percentage of the resources such as water, electricity, and manpower. This approach demands more time, money, and effort. The state of the art drip irrigation methods make use of the solenoid valve to control the water flow. The problem with such a system is reflected in its power consumption which is a significant factor for large-scale demands. The method proposed in this paper addresses this problem by developing an automated drip irrigation system that replaces components used in conventional methods with its economical counterparts in the market. A system using Node MCU, DC submersible motor, and soil moisture sensor is developed to automate the irrigation process ensuring efficient water and energy consumption. Since the proposed system utilizes economically cheaper components, it provides an upper edge over other systems in terms of expenditure and in turn economically feasible for large-scale demands. A mobile application is also developed to control, monitor, and schedule irrigation processes. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Enhancing Network Security with Comparative Study of Machine Learning Algorithms for Intrusion Detection
With the ever-increasing network systems and dependency on digital technologies, ensuring the security and integrity of these systems is of paramount importance. Intrusion detection systems (IDS) play a major role in sheltering such systems. Intrusion detection systems are technologies that are designed to monitor network and system activities and detect suspicious, unauthorized, malicious behavior. This research paper conducts a comprehensive comparative analysis of three popular machine learning algorithmsK-Nearest Neighbors (KNN), Random Forest (RF), and Logistic Regression (LR)in the context of intrusion detection using the renowned NSL-KDD dataset. Preprocessing techniques are applied, and the dataset is split for rigorous evaluation. The findings of this research highlight the effectiveness of Random Forest in detecting intrusions, showcasing its potential for real-world network security applications. This study contributes to the field of intrusion detection and offers valuable insights for network administrators and cybersecurity professionals to enhance network protection. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Analysis of Multinomial Classification for Legal Document Categorization
A major area of research today is the application of Machine Learning Techniques for Document or Text Classification. Document Classification is an important aspect of Electronic Discovery in the Legal domain. The need for the process to be automated has been realized over the past few years. Multinomial Classification is a well-known Supervised Machine Learning Technique that helps us classify if there are more than two classes used for the purpose of Classification. Evaluation metrics such as Precision, Recall, and F1 Score have been used to measure the efficiency of Classification. Logistic Regression and Gradient Boosting Algorithms have outperformed other Multiclass Classification techniques. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Cognitive outcomes prediction in children using machine learning and big data analytics
This study explores the potential of machine learning (ML) and big data analytics in predicting cognitive outcomes in children, aiming to enhance early identification and intervention strategies. Leveraging a diverse dataset comprising neurocognitive assessments, genetic markers, socio-economic factors, and environmental variables, our research employs advanced ML algorithms to develop predictive models. The interdisciplinary approach integrates neuroscience, psychology, and data science to discern patterns and correlations within the expansive dataset. The study emphasizes the importance of early cognitive assessment for optimal child development and academic success. By harnessing the power of big data, our models seek to uncover nuanced relationships that traditional methodologies may overlook. Preliminary results indicate promising accuracy in predicting cognitive outcomes, offering a valuable tool for educators, healthcare professionals, and policymakers. Additionally, the model's interpretability allows for a deeper understanding of the factors influencing cognitive development. Ethical considerations, privacy safeguards, and data governance are integral components of this research, ensuring responsible use of sensitive information. The implications of this study extend beyond academia, with the potential to inform educational policies, personalized learning strategies, and targeted interventions for at-risk populations. As technological advancements continue, the integration of ML and big data analytics in predicting cognitive outcomes heralds a new era in pediatric research, promoting proactive approaches to support children's cognitive well-being. 2024 IEEE. -
AI Based Seamless Vehicle License Plate Recognition Using Raspberry Pi Technology
This research presents the implementation of an innovative Vehicle Management System designed specifically for the Christ University Project 'CampusWheels.' The system incorporates cutting-edge technologies, including YOLOv8 and Tesseract OCR, for robust license plate recognition. Addressing the unique challenges faced by Christ University in managing and securing vehicular movements within the campus, this project becomes crucial as the number of vehicles on campuses continues to grow. It not only provides an effective solution to these challenges but also introduces innovative methodologies, marking a significant departure from conventional campus management practices. The paramount importance of this project lies in its ability to enhance campus security through real-time vehicle monitoring and identification. The utilization of YOLOv8 for vehicle detection and Tesseract OCR for license plate recognition ensures a high level of accuracy in identifying and tracking vehicles entering and leaving the campus. This precision significantly contributes to the prevention of unauthorized vehicle access, a common security concern on educational campuses. Moreover, the system's ability to streamline traffic flow and improve efficiency in parking and access control addresses practical issues faced by campus administrators and security personnel. 2024 IEEE.