Reliable Machine Learning Classifiers and Big Data Sentiment Analysis for Evaluating the Patient Health Care Opinion Systems

Abstract

The patients’ experience is considered a dominant reputation in the hospital administration and medical fields. Online patient reviews are recognized as an important criterion for evaluating hospital service quality and performance. The classical approach of evaluating service excellence is often found to be tedious. But with machine learning classifiers and opinion mining techniques the data assessing, and evaluation is made casual and its saves time. Currently, patient satisfaction and quality of service for patients in hospitals plays a major role in health care sector. This is accomplished by predicting the varied hidden patterns along with identifying the key components responsible for patient satisfaction. This research work tried to ensure patient satisfaction and quality of service by proposing Machine Learning Classifiers and Big Data Analytics for Evaluating the Patient Health Care Opinion Systems. Firstly, for doing sentiment analysis in patient service satisfaction, a systematic overview is presented which focuses on review comments and opinion polls related to the health sector in quality service, characteristics associated with patient satisfaction, comments on drug reviews and recommendations. Secondly, this proposed work provides novel SCSP Ensemble Model to Analyze Patient Health Care Opinion Systems. The Classification models are used to classify patients’ feelings as positive, negative, or neutral using a machine learning approach to predict superlative models in data analysis. Ensemble techniques are used to analyze the opinions classified by the model, and the recommendation for health care is analyzed based on sentiment polarity. The very reason behind selecting this topic is to provide a sound information system to the healthcare industry based on the tweets posted. Then this research work provides the novel feature selection method for identifying the key feature in Home Health Care Services using patient satisfaction data. In this, we examined the several components to evaluate the superiority of different health care services based on diverse metrics. Various machine learning algorithms are applied to guess the significant aspects affecting patient healthcare satisfaction. As recognized, the patient experience is an indispensable for assessing the quality in healthcare services. Next, this proposed research work illustrates novel Sentiment Analysis Approach for Evaluating the Patient Drug Satisfaction using the recommendation method for medicine by explaining the symptoms. Then, the performance of the proposed approaches is compared with the existing systems, and the Justification of Proposed approaches are represented. Our proposed framework shows the performance of 95% AUC accuracy and the overall accuracy of 93% in analyzing the social media reviews in the healthcare sector. It provides a recommendation percentage in the healthcare sector as per positive reviews. Also, this work explains the results analysis of the model proposed, and tenders the model verification by validating the models used for this work. Finally, the overall conclusion of proposed work is narrated with the novelty and the contributions. This chapter also gives the directions for future research works and open issues available in health care domain using the proposed approaches.