Customer churn, the discontinuation of services by existing customers, poses a significant challenge to the telecommunications industry. This paper proposes a novel adaptive ensemble learning framework for highly accurate customer churn prediction. The framework integrates multiple base models, including XGBoost, LightGBM, LSTM, a Multi-Layer Perceptron (MLP) neural network, and Support Vector Machine (SVM). These models are strategically combined using a stacking ensemble method, further enhanced by meta-feature generation from base model predictions. A rigorous data preprocessing pipeline, coupled with a multi-faceted feature engineering approach, optimizes model performance. The framework is evaluated on three publicly available telecom churn datasets, demonstrating substantial accuracy improvements over state-of-the-art techniques. The research achieves a remarkable 99.28% accuracy, signifying a major advancement in churn this http URL implications of this research for developing proactive customer retention strategies withinthe telecommunications industry are discussed.

In recent years, advancements in deep learning techniques have considerably enhanced the efficiency and accuracy of medical diagnostics. In this work, a novel approach using multi-task learning (MTL) for the simultaneous classification of lung sounds and lung diseases is proposed. Our proposed model leverages MTL with four different deep learning models such as 2D CNN, ResNet50, MobileNet and Densenet to extract relevant features from the lung sound recordings. The ICBHI 2017 Respiratory Sound Database was employed in the current study. The MTL for MobileNet model performed better than the other models considered, with an accuracy of74\% for lung sound analysis and 91\% for lung diseases classification. Results of the experimentation demonstrate the efficacy of our approach in classifying both lung sounds and lung diseases concurrently. In this study,using the demographic data of the patients from the database, risk level computation for Chronic Obstructive Pulmonary Disease is also carried out. For this computation, three machine learning algorithms namely Logistic Regression, SVM and Random Forest classifierswere employed. Among these ML algorithms, the Random Forest classifier had the highest accuracy of 92\%.This work helps in considerably reducing the physician's burden of not just diagnosing the pathology but also effectively communicating to the patient about the possible causes or outcomes.

Geometrically modified fiber optic sensors (FOS), particularly U-bent FOS, have gained significant attention due to their remarkably high refractive index (RI) and evanescent wave absorbance (EWA) sensitivity, as well as their ergonomic design and ease in handling. In this study, we present a theoretical model for the U-bent FOS probes, to predict the sensor behavior by numerically simulating the light propagation in an equivalent 2D semi-circular ring using ray tracing approach. In addition to the effects due to the modification of geometry, this study presents a thorough investigation of the influence of the bend-induced material deformation on the nature of light propagation and refractive losses. We introduce bend ratio (ratio of bend radius to fiber core radius), to explain the influence of geometry modification and the bend-induced inhomogeneity in RI (BIRI) of the fiber core on RI sensitivity. The bend ratio of bent plastic optical fiber sensors falls under one of the four bending regimes namely, gentle, geometric, saturation and plastic, for which the bend ratios are less than 35, 25, 17 and 7 respectively. The results also show that for bend ratios less than 7, BIRI inhomogeneity is responsible for the high RI sensitivity observed with U-bent probes as opposed to the simple geometric modification. This study also indicates the existence of an optimum bend ratio (for a given value of RI of the surrounding medium) where RI sensitivity is maximum. These findings were validated with previously reported experimental results.

With the advent of Internet of Things (IoT) and the increasing use of application-based processors, security infrastructure needs to be examined on some widely-used IoT hardware architectures. Applications in today's world are moving towards IoT concepts as this makes them fast, efficient, modular and future-proof. However, this leads to a greater security risk as IoT devices thrive in an ecosystem of co-existence and interconnection. As a result of these security risks, it is of utmost importance to test the existing cryptographic ciphers on such devices and determine if they are viable in terms of swiftness of execution time and memory consumption efficiency. It is also important to determine if there is a requirement to develop new lightweight cryptographic ciphers for these devices. This paper hopes to accomplish the above-mentioned objective by testing various encryption-decryption techniques on different IoT based devices and creating a comparison of execution speeds between these devices for a variety of different data sizes. Keywords-Internet of things(IoT), application-based processors, security, encryption-decryption, speed, efficiency

Cardiovascular diseases (CVDs) are a main cause of mortality globally, accounting for 31% of all deaths. This study involves a cardiovascular disease (CVD) dataset comprising 68,119 records to explore the influence of numerical (age, height, weight, blood pressure, BMI) and categorical gender, cholesterol, glucose, smoking, alcohol, activity) factors on CVD occurrence. We have performed statistical analyses, including t-tests, Chi-square tests, and ANOVA, to identify strong associations between CVD and elderly people, hypertension, higher weight, and abnormal cholesterol levels, while physical activity (a protective factor). A logistic regression model highlights age, blood pressure, and cholesterol as primary risk factors, with unexpected negative associations for smoking and alcohol, suggesting potential data issues. Model performance comparisons reveal CatBoost as the top performer with an accuracy of 0.734 and an ECE of 0.0064 and excels in probabilistic prediction (Brier score = 0.1824). Data challenges, including outliers and skewed distributions, indicate a need for improved preprocessing to enhance predictive reliability.

Hybrid energy storage systems (HESS) have carved a niche in the industry. HESS improve the system efficiency, reduce the overall cost and increase the lifespan of the system. The proton exchange membrane (PEM) fuel cell is hybridized with Li-ion batteries (LIB) for vehicular applications, robotic applications etc. In applications which have geometrical space constraints, the temperature of the energy storage elements is influenced by convective heat transfer. In this paper the thermal analysis of the geometry of PEM-LIB hybrid system is carried out using COMSOL Multiphysics Software package for different discharge rates (C rates) of the LIB and different voltages of the PEM cell. The additional rise in temperature of the LIB pack when placed in close proximity with PEM cell was in the range of 0.03-0.6$^0$C at 4C. The cell temperature of the LIB pack increased with increase in C rate and decrease in PEM cell voltage.