The effects of negative social media connotations on subjective wellbeing of an ageing population: A stressor-strain-outcome perspective.

In recent years, users' privacy concerns and reluctance to use have posed a challenge for the social media and wellbeing of its users. There is a paucity of research on elderly users' negative connotations of social media and the way these connotations contribute to developing passive behaviour towards social media use, which, in turn, affects subjective wellbeing. To address this research vacuum we employed the stressor-strain-outcome (SSO) approach to describe the evolution of passive social media use behaviour from the perspective of communication overload, complexity, and privacy. We conceptualized subjective wellbeing as a combination of three components-negative feelings, positive feelings, and life satisfaction. Negative and positive feelings were used to derive an overall affect balance score that fluctuates between 'unhappiest possible' and 'happiest possible'. The proposed research framework was empirically validated through 399 valid responses from elderly social media users. Our findings reveal that communication overload and complexity raise privacy concerns among social media users, which leads to passive usage of social media. This passive social media use improved the subjective wellbeing favourably by lowering negative feelings and raising positive feelings and life satisfaction. The findings also revealed that respondents' overall affect balance leans towards positive feelings as a consequence of passive social media use. This study contributes to the field of technostress by illuminating how the SSO perspective aid the comprehension of the way passive social media use influences the subjective wellbeing of its users.

Predicting dengue transmission rates by comparing different machine learning models with vector indices and meteorological data.

Machine learning algorithms (ML) are receiving a lot of attention in the development of predictive models for monitoring dengue transmission rates. Previous work has focused only on specific weather variables and algorithms, and there is still a need for a model that uses more variables and algorithms that have higher performance. In this study, we use vector indices and meteorological data as predictors to develop the ML models. We trained and validated seven ML algorithms, including an ensemble ML method, and compared their performance using the receiver operating characteristic (ROC) with the area under the curve (AUC), accuracy and F1 score. Our results show that an ensemble ML such as XG Boost, AdaBoost and Random Forest perform better than the logistics regression, Naïve Bayens, decision tree, and support vector machine (SVM), with XGBoost having the highest AUC, accuracy and F1 score. Analysis of the importance of the variables showed that the container index was the least important. By removing this variable, the ML models improved their performance by at least 6% in AUC and F1 score. Our result provides a framework for future studies on the use of predictive models in the development of an early warning system.

A protocol for developing a classification system of mosquitoes using transfer learning.

Mosquito identification and classification are the most important steps in a surveillance program of mosquito-borne diseases. With conventional approach of data collection, the process of sorting and classification are laborious and time-consuming. The advancement of computer vision with transfer learning provides excellent alternative to the challenge. Transfer learning is a type of machine learning that is viable and durable in image classification with limited training images. This protocol aims to develop step-by-step procedure in developing a classification system with transfer learning algorithm for mosquito, we demonstrate the protocol to classify two species of Aedes mosquito - Aedes aegypti L. and Aedes albopitus L, but user can adopt the protocol for higher number of species classification. We demonstrated the way of start from the scratch, fine-tuning two pre-trained model performance by using different combination of hyperparameters - batch size and learning rate, and explain the terminology in the Appendix. This protocol target on the domain expert such as entomologist and public health practices to develop their own model to solve the task of mosquito/insect classification.

Implementation of a deep learning model for automated classification of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse) in real time.

Classification of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse) by humans remains challenging. We proposed a highly accessible method to develop a deep learning (DL) model and implement the model for mosquito image classification by using hardware that could regulate the development process. In particular, we constructed a dataset with 4120 images of Aedes mosquitoes that were older than 12 days old and had common morphological features that disappeared, and we illustrated how to set up supervised deep convolutional neural networks (DCNNs) with hyperparameter adjustment. The model application was first conducted by deploying the model externally in real time on three different generations of mosquitoes, and the accuracy was compared with human expert performance. Our results showed that both the learning rate and epochs significantly affected the accuracy, and the best-performing hyperparameters achieved an accuracy of more than 98% at classifying mosquitoes, which showed no significant difference from human-level performance. We demonstrated the feasibility of the method to construct a model with the DCNN when deployed externally on mosquitoes in real time.