Search
10 articles for “ROC curve”
-
A Novel Hybrid Link Prediction Algorithm for E-Commerce Recommender System Based upon Common Neighbor and Resource Allocation Methods
Abstract: Link prediction is a challenging task in recommender systems, as it requires the ability to accurately predict future links between users and items. In this study, we propose a novel hybrid link prediction algorithm for e-commerce recommender systems that combines the common neighbor and resource allocation methods. The common neighbor method is a straightforward and intuitive algorithm that calculates the number of shared neighbors between two nodes. The intuition is …
Published in International Journal of Data Structure Studies · Vol. 1, Issue 1, 2023 · pp. 12–17 Read article
-
The Evaluation of the Light Parameters and Pleural Fluid Cholesterol to Determine the Differences Between Exudative and Transudative Pleural Discharge
Abstract: Background: Pleural effusion occurs when an imbalance between pleural fluid production and absorption leads to the accumulation of excess fluid in the pleural cavity. Pleural effusions are commonly classified into two types: transudative or exudative, depending on the underlying mechanism of fluid formation. While transudates typically result from systemic factors like heart failure or liver cirrhosis, exudates are usually caused by local factors such as infection, malignancy, or inflammation. Differentiating …
Published in Research and Reviews : A Journal of Life Sciences · Vol. 14, Issue 3, 2024 · pp. 30–35 Read article
-
A Comprehensive Review of CNN-Based Framework for Multi-Sign Detection of Diabetic Retinopathy in Fundus Images Using Public Datasets
Abstract: Diabetic retinopathy (DR) is one of the main causes of vision impairment. Blindness prevention and effective treatment depend on early detection. A thorough deep learning-based framework for the automatic segmentation and simultaneous detection of exudates, hemorrhages, and microaneurysms – three important DR indicators – from retinal fundus images is presented in this work. These three pathological signs’ corresponding annotated image patches, along with background (no-sign) areas, were used to train …
Published in Research and Reviews : Journal of Computational Biology · Vol. 14, Issue 3, 2025 · pp. 14–23 Read article
-
Integration of Biomarkers in III and IV CKD Patients for Evaluation of Their Predictive Value for CVD in CKD Patients
Abstract: Background: Chronic kidney disease (CKD) is presently characterized by the presence of proteinuria and/or a diagnosis of renal impairment. The worldwide significance of CKD is underscored by the fact that its prevalence and incidence have doubled over the past three decades. This paper provides an overview of the existing evidence regarding biomarkers in individuals with CVD or CKD, with a particular focus on the emerging biomarkers (i.e., eGFR, sALB, UACR, …
Published in Research and Reviews: A Journal of Medicine · Vol. 15, Issue 3, 2025 · pp. 1–12 Read article
-
AI-Driven Predictive Maintenance Framework for Intelligent Vehicle Health Monitoring
Abstract: The accelerated development of smart and connected car systems made the necessity to find the accurate and real-time predictive maintenance solutions which would minimize the number of unexpected failures as well as increase the cars on-road safety. The current paper proposes an artificial intelligence-based hybrid predictive maintenance system that combines Long Short-Memory (LSTM) networks and the XGBoost predictor to provide a potent vehicle fault diagnosis, Remaining Useful Life (RUL) prediction, …
Published in Trends in Machine design · Vol. 13, Issue 1, 2026 · pp. 1–17 Read article
-
Passive Digital Phenotyping for Longitudinal Burnout and Occupational Mental Health Surveillance: A Transformer-Based Explainable Deep Learning Approach Using Smartphone Behavioral Streams
Abstract: Occupational burnout constitutes a pervasive yet chronically under-surveilled public health threat, its insidious temporal evolution rendering episodic self-report instruments structurally inadequate for early detection. This paper introduces BurnoutSense, a passive digital phenotyping framework that continuously harvests eight heterogeneous smartphone behavioral data streams encompassing application usage ecology, communication metadata, geospatial mobility, screen interaction dynamics, inferred sleep rhythmicity, keystroke kinematics, ambient noise exposure, and battery/charging cadence to construct individualized multivariate behavioral signatures …
Published in International Journal of Behavioral Sciences · Vol. 3, Issue 2, 2026 · pp. 44–53 Read article
-
Early Pregnancy Levels of Fasting Glucose, HbA1c, and Adiponectin as Predictors of Gestational Diabetes Mellitus Among Pregnant Women in Tamil Nadu, India
Abstract: Background: Gestational diabetes mellitus (GDM) is rapidly becoming a major public health issue across India, and Tamil Nadu continues to report some of the country’s highest incidence figures. Identifying women at elevated risk during the first trimester allows health workers to intervene early and improve outcomes for both mothers and babies. This research, therefore, examines whether fasting plasma glucose, glycated haemoglobin, and adiponectin measured at that initial visit can reliably …
Published in Research and Reviews : A Journal of Medical Science and Technology · Vol. 15, Issue 2, 2026 · pp. 10–18 Read article
-
Prediction of Customer Churn Using Machine Learning Classification Models
Abstract: Customer churn prediction is a critical task in both the telecommunication and medical industries, where retaining customers or patients is essential for ensuring long-term profitability and maintaining high-quality service. To address this, a range of machine learning models—including logistic regression, decision trees, random forests, gradient boosting machines, and support vector machines—were employed to accurately forecast churn behavior. Prior to model training, the dataset underwent thorough preprocessing, which included handling missing …
Published in Journal of Computer Technology & Applications · Vol. 16, Issue 2, 2025 · pp. 86–92 Read article
-
Detecting Fake Accounts on Social Media Using Machine Learning
Abstract: The growing frequency of fake accounts on social media platforms underscores the critical necessity for effective detection methods. In response to this challenge, our study leverages state-of-the-art machine learning techniques to identify and counter deceptive entities effectively. By conducting a thorough analysis of social media data, our approach unveils intricate patterns indicative of fraudulent accounts, enabling proactive measures against them. Through the application of advanced algorithms, we present a comprehensive …
Published in Journal of Mobile Computing, Communications & Mobile Networks · Vol. 11, Issue 2, 2024 · pp. 21–32 Read article
-
Comparison of Several Clinical Scoring Systems in Predicting the Outcome of Variceal Bleeding
Abstract: Background: Stratification of variceal bleeding patients into high-risk and low-risk group is very important to guide them through the suitable clinical pathway and to save the medical costs. We purposed to find out the best scoring system in the prediction of rebleeding and death after variceal bleeding by comparing four clinical scoring systems(clinical Rockall score, complete Rockall score, AIMS65 score, Child-Pugh score) which seemed to be applicable and simple. Method: …
Published in Research and Reviews: A Journal of Medicine · Vol. 15, Issue 2, 2025 Read article