The user's feedback on healthcare services is usually based on ratings from post-service questionnaires. However, in order to get a clear view of the user's perspective, online text reviews need to be analyzed. We combined targeted and aspect-based sentiment analysis by multi-level attention to get a specific user sentiment on a target of an aspect. The multi-level attention consists of Target-level and Sentence-level attention. Our proposed framework is based on Bidirectional Gated Recurrent Unit. Bi-GRU is commonly known to have comparable results compared to LSTM while having lesser computational complexity. We also utilized Bidirectional LSTM based Character-Enhanced Token-Embedding to handle out of vocabulary words and misspelling to avoid error in detecting sentiment. We created a dataset of online healthcare reviews from 2018-2020, targeting the name of the hospital or department, with ten aspects: cleanliness, cost, doctor, food, nurse, parking, receptionist and billing, safety, test and examination, and waiting time. To improve the results of our proposed method, we calculated polarity weight to handle imbalanced aspects in the dataset. We classified these reviews into three polarities, which are positive, negative, and neutral. Based on our experiments, we achieved the best F1-Score of 88%.
Campaigns on online social media are becoming intriguing because all parties can compete with each other. Various content can be created to support or defeat one another. Interaction is important to control the battlefield in the face of disinformation. The campaign team must be able to increase interaction and reporting it as a form of achievement. The reality is that the campaign team has not been able to provide a measurable report on the success rate of political campaign interaction on social media. This is because there is no standardization in measuring these interactions. This fact inspires to build a model for measuring campaign interaction on social media. This model will be able to help the campaign team. This study uses a dataset from a governor election campaign in Indonesia which is then tested on a presidential election campaign dataset in America. The model was built using machine learning with the k classification method Nearest Neighbor (kNN) and Distance Weight kNN (DWkNN). The research stages were arranged using the Nassi-Shneiderman Framework. This framework describes the stages and comparison of the use of the KNN and the DWkNN from the training to testing stages in an easy to understand manner. The kNN and DWkNN training stages showed excellent accuracy results of nearly 100%. Furthermore, at the testing stage using the crossvalidation method with a variety of fold 5 to 20 variations showed excellent results, namely 99.89% on the governor election dataset and a range of 98% on the American presidential election dataset. This classification model has been tested using several datasets from several candidate social media accounts. We also created a dataset campaign interaction. This study shows that the proposed model can outperform previous studies. So, that it provides novelty in the form of a campaign interaction guide model for the campaign team.
We have developed an interactive statistical analysis system (ISAS-Q) with which clinical investigators with little experience in computers and programming can easily perform statistical analyses. ISAS-Q can perform most of the frequently used statistical methods, including multivariate analysis, in an interactive mode. Furthermore, ISAS-Q has self-consistent and extensive help functions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.