🏫 Student Academic Status Prediction
In this problem, you must implement a classification model capable of predicting a student's academic status (Target) using a provided dataset. The dataset is organized in a CSV file, and the model's performance will be evaluated using the F1-score.
🔹 Dataset
The dataset contains the following columns:
- SampleID: Unique identifier for each student
- Marital status: Student’s marital status (Single, Married, Other)
- Application mode: Application method (Online, In person, Other)
- Application order: Order of application (numeric)
- Course: The course the student enrolled in
- Daytime/evening attendance: Attendance type (Daytime/Evening)
- Previous qualification: Previous education level (High School, College)
- Nacionality: Student's nationality
- Mother's qualification: Mother's education level
- Father's qualification: Father's education level
- Mother's occupation: Mother's occupation
- Father's occupation: Father's occupation
- Displaced: Whether the student is displaced/away from home (Yes/No)
- Educational special needs: Special educational needs (Yes/No)
- Debtor: Whether the student has unpaid fees (Yes/No)
- Tuition fees up to date: Whether tuition fees are up to date (Yes/No)
- Gender: Student’s gender (Male/Female)
- Scholarship holder: Whether the student receives a scholarship (Yes/No)
- Age at enrollment: Age when enrolling
- International: International student status (Yes/No)
- Curricular units 1st sem (credited): Credited units in the first semester
- Curricular units 1st sem (enrolled): Enrolled units in the first semester
- Curricular units 1st sem (evaluations): Evaluations completed in the first semester
- Curricular units 1st sem (approved): Approved units in the first semester
- Curricular units 1st sem (grade): Grades obtained in the first semester
- Curricular units 1st sem (without evaluations): Units without evaluations in the first semester
- Curricular units 2nd sem (credited): Credited units in the second semester
- Curricular units 2nd sem (enrolled): Enrolled units in the second semester
- Curricular units 2nd sem (evaluations): Evaluations completed in the second semester
- Curricular units 2nd sem (approved): Approved units in the second semester
- Curricular units 2nd sem (grade): Grades obtained in the second semester
- Curricular units 2nd sem (without evaluations): Units without evaluations in the second semester
- Unemployment rate: Unemployment rate in the student’s region
- Inflation rate: Inflation rate
- GDP: Gross Domestic Product
- Target: The target variable, representing the student's academic status (Dropout, Enrolled, Graduate)
📊 Dataset Notes
- Target field:
Target
- The model must be trained on
train.csv and evaluated on test.csv.
⚙️ Evaluation Criteria
- Performance Metric: F1-score (the higher, the better)
📨 Submission Format
The submission file must be a CSV with exactly two columns:
| Column | Type | Description |
|---|
SampleID | integer | Unique identifier for each row in the test set |
Target | string | Predicted class for the student (Dropout, Enrolled, Graduate) |
🔹 Example
| SampleID | Target |
|---|
| 101 | Dropout |
| 102 | Enrolled |
| 103 | Graduate |
Important: Each row in the CSV represents the prediction for one student in the test set.
📝 Note
The score is based on the F1-score obtained by the model on the test set.
To achieve maximum points, the F1-score must be at least 0.75, while values below 0.5 receive 0 points.
Intermediate values are scaled proportionally.
🗂️ Useful Resources
