An Artificial Intelligence-Based Reactive Health Care System for Emotion Detections

Abstract

In the past few years, remote monitoring technologies have grown increasingly important in the delivery of healthcare. According to healthcare professionals, a variety of factors influence the public perception of connected healthcare systems in a variety of ways. First and foremost, wearable technology in healthcare must establish better bonds with the individuals who will be using them. The emotional reactions of patients to obtaining remote healthcare services may be of interest to healthcare practitioners if they are given the opportunity to investigate them. In this study, we develop an artificial intelligence-based classification system that aims to detect the emotions from the input data using metaheuristic feature selection and machine learning classification. The proposed model is made to undergo series of steps involving preprocessing, feature selection, and classification. The simulation is conducted to test the efficacy of the model on various features present in a dataset. The results of simulation show that the proposed model is effective enough to classify the emotions from the input dataset than other existing methods.

Improving Healthcare Communication: AI-Driven Emotion Classification in Imbalanced Patient Text Data with Explainable Models

“…Table 4 highlights that the classification results obtained by this BERT and LSTM models outperformed the MultiLayer perceptron implemented by Azam et al [4]. Although these results are satisfactory, they remain lower than those of the intelligent water drop algorithm based on the back-propagation neural network (BPNN) model [14]. It seems obvious that using the intelligent water drop algorithm for feature selection optimization made a difference and significantly improved the classification result.…”

“…Encouraged by these early successes, subsequent research has sought to push the boundaries of sentiment analysis in healthcare even further. In Mohammad et al [14], for example, the intelligent water drop algorithm was used to select informative features from EmoHD, demonstrating the potential of innovative algorithms to improve feature selection and, therefore, feelings of classification accuracy in health-related texts.…”

“…By leveraging insights from the EmoHD dataset, this study (in comparison with [4,14]) aims to contribute further by evaluating the performance of different learning algorithms such as logistic regression, Bayes naive, and LightGBM. integrating a variety of metrics (CBOW, TF-IDF, and Word2vec) and comparing them to recurrent deep learning networks, including LSTM and BERT models which have shown good results in sentiment classification [10][11]13].…”

Ontological Modeling for Contextual Data Describing Signals Obtained From Electrodermal Activity for Emotion Recognition and Analysis

“…T HE recognition of human emotions by computer systems is finding applications in a growing number of fields [34], such as distance learning [75,19], healthcare [20,40,65], marketing [52,43] and many others [8,23]. Depending on the availability of particular channels, recognition methods can use different signals as a source of insight into human emotions.…”