AI-Driven Real-Time Feedback System for Enhanced Student Support: Leveraging Sentiment Analysis and Machine Learning Algorithms

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2024, Номер 10(4)

Опубликована: Дек. 21, 2024

The rapid evolution of educational technologies has led to a shift toward personalized and adaptive learning experiences. A critical component such systems is the ability provide timely relevant feedback students. This paper presents an AI-driven real-time system designed enhance student support through integration sentiment analysis machine algorithms. leverages gauge emotional tone interactions, as forum posts, assignment submissions, feedback. Machine algorithms, including decision trees, vector machines (SVM), deep models, are used analyze predict engagement, performance, states. By combining both cognitive insights, delivers personalized, context-sensitive that helps students overcome challenges improve academic outcomes. effectiveness evaluated using multiple datasets, showing significant improvements in satisfaction, performance.

Язык: Английский

---

Optimizing Type II Diabetes Prediction Through Hybrid Big Data Analytics and H-SMOTE Tree Methodology

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Янв. 10, 2025

In the last few years, Type II diabetes has become much more common worldwide, presenting major problems for both healthcare systems and individuals. Utilizing big data analytics shown potential as a means of forecasting managing persistent illnesses, like diabetes. This paper proposes novel hybrid approach that combines techniques with an H-SMOTE tree algorithm prediction The suggested method addresses class imbalance present in medical datasets improves accuracy by combining steps feature selection, preprocessing, classification. order to prepare raw analysis, it must first be cleaned, standardised, transformed. Then, selection are used identify most important factors help predict streamlines predictive model lowers its dimensionality. classification phase, called is used. two existing techniques: Hoeffding Adaptive Tree (HAT) Synthetic Minority Oversampling Technique (SMOTE). tackles imbalanced creating synthetic samples under-represented class, while also adapting decision structure receives new data. Experiments show this effective accurately predicting researchers found outperformed other machine learning methods, classic recent ones. words, was accurate T2DM cases. evident terms several metrics, including how well identified true positives (sensitivity), avoided false (specificity), overall performance captured AUC-ROC score. Additionally, proposed displays resilience scalability, rendering apt extensive frequently encountered within domains.

Язык: Английский

---

Rainfall Forecasting in India Using Combined Machine Learning Approach and Soft Computing Techniques : A HYBRID MODEL

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Янв. 9, 2025

Accurate rainfall prediction in India is crucial for agriculture, water management, and disaster preparedness, particularly due to the reliance on southwest monsoon. This paper examines historical trends from 1901 2022, highlighting significant anomalies changes identified through Pettitt test. The effectiveness of advanced machine learning techniques explored Artificial Neural Network-Multilayer Perceptron (ANN-MLP) enhancing forecasting accuracy compared with statistical methods. By integrating important climate variables—temperature, humidity, wind speed, precipitation into ANN-MLP model, its ability capture complex nonlinear relationships demonstrated. Additionally, analysis employs geo-statistical techniques, specifically Kriging, visualize spatial-temporal variability across different regions India. findings emphasize potential modern computational methods overcome traditional challenges, ultimately improving decision-making agricultural planning resource management face variability.

Язык: Английский

---

A Systematic Comparative Study on the use of Machine Learning Techniques to Predict Lung Cancer and its Metastasis to the Liver: LCLM-Predictor Model

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Янв. 11, 2025

Lung cancer is one of the major causes deaths with thousands affected patients who have developed liver metastasis, complicating treatment and further prognosis. Early predictions lung metastasis may greatly improve patient outcomes since clinical interventions will be instituted in time. This paper compares performance different machine learning models including Decision Tree Classifiers, Logistic Regression, Naïve Bayes, K-Nearest Neighbors, Support Vector Machines Gaussian Mixture Models toward best set techniques for prediction. The applied dataset includes various features, such as respiratory symptoms biochemical markers, development stronger predictive performance. were cross-validated using testing validation aimed at generalizing whole model reliability generating both train test data. results generated are gauged metrics accuracy, precision, recall, F1-score, area under ROC curve. Results obtained revealed that KNN also showed accuracy strong classification performance, especially early-stage metastasis. present study a comparison models, which hence denotes potential these decision-making suggests application to diagnostic tools early detection cancer. provides very useful guide applicable use oncology helps pave way future research would focused on optimization integration into healthcare systems produce better management survival rates.

Язык: Английский

---

CBDC-Net: Recurrent Bidirectional LSTM Neural Networks Based Cyberbullying Detection with Synonym-Level N-Gram and TSR-SCSOFeatures

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2024, Номер 10(4)

Опубликована: Дек. 20, 2024

Social networks Cyber bullying has become another common problem in online social (OSNs) which exposes individuals to high risks of their mental health and interacting with others. Previous work cyber detection is often confronted limitations accurately detecting abusive behavior because the intricacies space evolution practices. A new approach classification network (CBDC- Net) for improving effectiveness OSNs based on natural language processing features, feature selection techniques, deep learning algorithms also presented this study. CBDC-Net can overcome these challenges existing methods using innovative Natural Language Processing (NLP) Deep Learning approaches. In data preprocessing step, filter normalize text that openly collected from OSNs. After that, extracts features a Synonym Level N-Gram (SLNG) it incorporates both word character-based information make synonyms much better than other method. CSI applied Textual Similarity Resilient Sand Cat Swarm Optimization (TSR-SCSO) give an iterative value features’ importance level detect bullying. Last, CBDC-Net, Recurrent Bidirectional Long Short-Term Memory (LSTM)Neural Network (RBLNN) used as applied, recognizes sequential nature textual enabling proper distinction between cases. Last but not least, CBDC Net provides promising solution solving mentioned problems

Язык: Английский

---

Comparative Evaluation of Feature Selection Techniques and Machine Learning Algorithms for Alzheimer's Disease Staging

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(2)

Опубликована: Март 21, 2025

Dementia encompasses a range of brain disorders characterized by cognitive decline, with memory loss as hallmark symptom. Alzheimer's disease (AD), the most common form dementia, progressively affects functions, leading to severe loss. Early and accurate detection AD is essential for timely intervention, preventing further neuronal damage, improving patient outcomes. This study employs machine learning (ML) techniques, feature selection methods, texture analysis enhance diagnosis. By systematically evaluating various techniques Principal Component Analysis (PCA) in conjunction multiple ML algorithms, identifies effective approach classifying stages. The integration texture-based features models demonstrates significant improvement distinguishing Cognitive Normal, Mild Impairment, These findings highlight clinical significance combining early diagnosis, facilitating more precise classification contributing personalized treatment strategies.

Язык: Английский

---

Survey on Resume Parsing Models for JOBCONNECT+: Enhancing Recruitment Efficiency using Natural language processing and Machine Learning

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2024, Номер 10(4)

Опубликована: Дек. 15, 2024

Due to the rapid rise of digital recruitment platforms, accurate and fast resume processing is needed speed hiring. JOBCONNECT+-specific algorithms improvements are extensively covered in investigation. Better parsing technologies may reduce candidate screening time resources, which this survey encourage. Despite breakthroughs Natural language Machine Learning (NLP ML), present fail extract categorise data from different forms, hindering recruiting. The Multi-Label Parser Entity Recognition Model (M-LPERM) employs entity recognition multi-label classification increase accuracy flexibility handle explosion complexity modern formats. adaptable approach satisfies JOBCONNECT+ criteria handles formats with varying language, structure, content. Automatic shortlisting, skill gap analysis, customised job suggestions included research. In a complete simulation examination, M-LPERM compared existing models for accuracy, speed, format adaptability.

Язык: Английский

---