Exploring Academic Success: Visualizing Student Performance and Attendance using UCI Machine Learning Repository: Predict students' dropout and academic success.
The notebook was created based on a dataset of student information from a higher education institution regarding utilising machine learning methods to identify academic performance and the remedy to fight dropout risks at an early stage of a student's academic path. This data could be utilized by higher education institutions to gain a better understanding of their students' academic progress and pinpoint areas for improvement from both individual and institutional standpoints. This notebook attempted to discover any recognizable pattern for students' educational accomplishment concerning the time of attendance and communicate the findings using descriptive statistics and graphs.
- The graduates take up half of the dataset while the dropouts takes 32% of the dataset's population.
- Most of individuals are from 18 to 30, with some of the older individual as far as 70.
- The distribution of students across grades suggests that evening attendance students may perform worse than daytime attendance students.
- A linear correlation in students' performance between the two semesters was found, with most students retaining their performance from the 1st to the 2nd semester regardless of the number of courses enrolled.
Environment:
- Numpy 1.20.3
- Pandas 2.0.0
- Matplotlib 3.4.2
- Sklearn 1.0.2
References: Predict students' dropout and academic success. UCI Machine Learning Repository