This paper presents an approach for detecting professional vs. non-professional written healthcare text using language models. To this end, we first collected a data set containing articles crawled from the internet, then based on the source website tagged those articles into two categories, namely professional and non-professional. We then trained two models for text classification, using Naive-Bayes-based and RoBERTa-based methods, respectively. Our models achieved remarkable results, with an accuracy of over 98 % on the test data set, as evaluated through various metrics. Our findings provide important insights into the effective use of language models to detect the professional level of healthcare articles.
This study proposes a technique for improving identity verification at checkpoint units utilizing cloud-based face recognition technology, which involves a tablet PC with a webcam and a network as a checkpoint station. The system captures video images using WebRTC, while OpenCV-JS is a version specifically designed to function on web browsers that support WASM. The HAAR Cascade Classifier will be used to detect faces inside preprocessed images and post them to the server side through an API. The API will also recognize faces using Amazon Rekognition, resulting in a face ID to look for participant information from the MariaDB database. Participants authenticate themselves and log in; a message is sent to other relevant systems over MQTT. Before accessing the system, user permissions will be granted by integrating with OAuth 2.0 to preserve security. Data on attendance events is gathered and visualized in Power BI dashboards. 126,183 samples, from 2019 to 2024, were used for system evaluation. The system has been utilized for several purposes, including verifying attendance at graduation ceremonies, requesting rights at the COVID-19 vaccination station, and verifying exam admission. The experiment revealed that the new approach could reduce the 33-second waiting time at the checkpoint to only 3 seconds and, at the longest, no more than 7 seconds per attendee. The result is equivalent to a 91 % gain in efficiency.
On behalf of Prince of Songkla University (PSU), I would like to invite you to The 21st International Joint Conferenc on Computer Science and Software Engineering (JCSSE 2024) is co-organized by IEEE Thailand Section (IEEE Computer Society Thailand Chapter) and College of Computing, Prince of Songkla University.
This study examines the implementation of collaborative learning in Basic Physics education through the utilization of Moodle Learning Management System (LMS). The investigation focuses on the impact of LMS integration on group discussions and problem-solving activities. Methods involved the random formation of 17 groups, consisting of 51 students, who engaged in both face-to-face and LMS-supported discussions. Group assessments were conducted, comparing pre-LMS and post-LMS usage scores, using two-tailed t-test analysis to determine differences in student performance. Results indicate that the incorporation of Moodle LMS in the collaborative learning process positively influenced group discussions and problem-solving outcomes, as evidenced by significant differences between pre-LMS and post-LMS scores. These findings highlight the potential of LMS integration to enhance collaborative learning experiences and improve student engagement in Basic Physics education.
Effective hospital resource management is crucial for service cost and workforce efficiency, especially during abnormal situations with sudden changes in patient volumes. Hospitals often face challenges in resource management and find it difficult to predict patient numbers due to uncertain and volatile data. If a hospital can accurately predict the number of patients during abnormal situations, it can allocate resources to align with the expected patient influx. This helps the hospital maintain appropriate costs and optimize workforce efficiency. In this research, we propose a solution to predict the number of daily outpatient visits from a private hospital in Thailand during the COVID-19 outbreak period as a case study. We use the Ensemble Empirical Mode Decomposition (EEMD) to preprocess the data by decomposing it into intrinsic mode functions, reducing the impact of data volatility and predict these intrinsic mode functions using a Long Short-Term Memory (LSTM) model with the service utilization pattern (seasonal value) as a lookback period to enhance prediction efficiency. This approach is referred to as the EEMD-LSTM model. Experimental results indicate that, compared to LSTM, ARIMA, and AutoML models evaluated for prediction periods of 7, 14, 21, and 28 days using metrics including MAE, RMSE, MAPE, and MAE Ratio, the EEMD-LSTM model demonstrates superior performance in predicting patient numbers during abnormal events. Unlike other models, the EEMD-LSTM model exhibits minimal changes in error when the prediction period is adjusted, highlighting its robustness. Overall, the EEMD-LSTM model consistently outperforms all other models assessed. When applied to data from the orthopedic department, although the MAE values are similar, noticeable differences in trends are observed. Moreover, when predicting data from the pediatric department, the EEMD-LSTM model proves to be more effective than all other models. This model can be effectively utilized by hospitals for resource management and planning, including medication and staffing, ensuring optimal service delivery to patients during abnormal situations.
