Student Performance Dataset

Proven hypothesis

The academic performance of a student depends not only on their academic capabilities but also their socio-economic status.

Imagine this...

You are a headmaster/headmistress, managing a school of thousands of students. How do you create the ideal environment for students to academically excel in their education?

ET VOILA!

The Student Forecaster Model:

https://student-forecaster.streamlit.app/

Step 1: Exploratory analysis

• Distribution and spread of the data
• Dropping columns which at first look are not relevant

Step 2: Creating subgroups of data

• To map out relationships between columns of data

Step 3: Analysis within sub-groups

• To decide the dominating variables most explaining the target variable

Summary of analytical methods used:

• Heatmap correlation
• Value counts
• Grouping by bins
• Histograms
• Boxplot

Conclusion from data analysis

• The biggest impact to G3 is failure
• If a student has failed then they likely have additional support
• Nearly 95% of students want to pursue higher education, so data is heavily skewed
• Over 90% of students do not have paid support but 60% of students have support at home

Machine Learning:

Preprocessing, Modelling, Scoring

Classification model used: Gradient Boosting Classifier

The classification model chosen:

Gradient Boosting Classifier

Model Scoring Metrics

print("Accuracy:",round(accuracy,2))
print("Precision:", round(precision,2))
print("Recall:", round(recall,2))
print("F1 Score:", round(f1,2))

Accuracy: 0.77
Precision: 0.76
Recall: 0.9
F1 Score: 0.82

Thank you for your attention