📈 CS Student Strength Analysis

This project explores the relationship between academic performance and problem-solving ability among Computer Science students. Using data visualizations and statistical insights, we break down how performance in core CS courses correlates with the number of coding problems solved — a key indicator of applied skill.

---

🧠 Objective

To analyze whether higher academic performance (GPA) in technical subjects such as OOP, Algorithms, and AI correlates with problem-solving strength, and to identify which subjects show the most variance or consistency among students.

---

📁 Dataset

• CS Student Strength.csv
• Includes GPA data from 20+ CS courses (theory and sessional)
• Also includes "Total Problems Solved", indicating applied coding skill

---

📊 Visualizations

1. 📊 Average GPA Per Course

Bar plot showing mean GPA across all subjects
✅ Conclusion: Students performed best in DBMS, Artificial Intelligence, and Discrete Math. Subjects like Machine Learning and Algorithm showed lower average GPAs.

---

2. 🧪 Grade Distribution in Core CS Subjects

Box plot of GPA distribution in key subjects
✅ Conclusion: Subjects like Data Structure, Operating Systems, and OOP had a wider GPA range, suggesting varying student comprehension. Compiler Design showed a tighter distribution.

---

3. 📈 Problem-Solving vs Course GPA

Scatter plots comparing "Total Problems Solved" with GPA in:

• OOP
• Algorithm
• Artificial Intelligence

✅ Conclusion: There appears to be a positive correlation — students with higher GPAs in these subjects tend to solve more problems. This supports the idea that academic performance and applied skill are linked, though not perfectly.

---

4. 🔥 Correlation Heatmap

Heatmap of correlations between all subjects and total problems solved
✅ Conclusion:

• Moderate correlation between Artificial Intelligence, Algorithm, and problem-solving.
• Some theory-heavy subjects (like Theory of Computing) showed weaker correlations, indicating that not all GPA scores predict practical ability.

---

📌 Key Insights

• Students who perform well in AI, Algorithms, and OOP tend to solve more coding problems.
• Some students excel at problem-solving even if they struggle in theory-heavy courses.
• There’s a strong need to balance theoretical and applied learning in CS education.

---

🧰 Built With

• Python 3
• pandas, NumPy
• matplotlib, seaborn

---

🧠 What I Learned

• How to clean and analyze structured academic data
• How to visualize relationships between variables
• That GPA is useful, but not always the best predictor of applied CS strength

---

---

👨‍💻 Author

Emdya Permuy-Llovio