LEARNING ANALYTICS TOOL FOR STUDYING THE CONDITIONS IN ROMANIAN MATERNITY HOSPITALS
University Politehnica of Bucharest (ROMANIA)
The management of maternity hospitals is a key element which is studied at medical universities. The proposed online platform analyzes using Elasticsearch the acquired feedback data from various Romanian maternity hospitals that is stored in the MongoDB NoSQL database. When new data is added by the medical students, a new document is added inside the MongoDB collection and the characteristics used for matching are indexed inside Elasticsearch. Regular expressions are utilized for extracting the relevant information. While performing the crawling, the features regarding the management characteristics like location, number of beds, intensive care units, staff qualifications, equipment and medicine availability are found using XPath to determine the relationships between the features. The important features that are clustered are detected by creating a document term matrix based on term frequency-inverse document frequency. The matching is done according to the characteristics by using the weighted terms according to the matrix. After the matching is done, the students, as well as the pregnant women have access to the best case situation with the finest management which is suggested based on the previous analyzed cases. The medical students collaborate and improve the existent knowledge base of the system after being approved by their professors, as well as they can learn how to manage better manage a maternity hospital according to the automatic assignment which is done for determining the best suited characteristics. After a period of time, feedback comes from the patients based on the implemented solution and the students, as well as the professor can use it for learning. The shift to the online platform has provided a close to reality image of today's maternity hospitals, while the automatic management recommendation aims to enhance the facilities. The pregnant women, the medical students and their professors learn from the previous successfully tested cases.