| dc.description.abstract | Universities currently face a significant management problem. The number of students
sharply grows that drags on the increase of the number of students dropping out of school every
year. This study aims to determine why the students drop out of school and give a solution to
detect the student cases who tend to give up studying to prevent or support them. In this context,
the identification signal of warning cases is defined as the signal groups. The students who have
more signals will be highly warned to tend dropping out of school.
To test the hypothesis, the content is minimized in the data of a faculty. At the first phase,
data is randomly generated as the historical records. Those records are normalized into
meaningful form which is the signal. Based on the available data, we can find a way to calculate
the levels of students dropping out of school risk called as the warning point. As a result, the
higher warning point students get, the higher risk that they tend to give up studying. To clarify
this statement, a visualization dashboard extremely supports for human’s forecast. In this test
case, it is also reviewed by the implementation.
This proposal system slightly contributes to the school management process. In the future,
this system could be integrated with other modern technologies such as data warehouse and
Kafka to have higher accuracy for the detection. | en_US |
