📘 Computer Data Analysis – Project Summary

This repository contains the full implementation of three course projects completed as part of the subject Computer Data Analysis. The projects are based on the well-known Iris dataset and focus on the practical application of data exploration, visualization, and machine learning techniques using Python.

🧩 Project Overview

✅ Project 1 – Exploratory Data Analysis

• Loaded and preprocessed the dataset.
• Calculated key descriptive statistics (mean, median, min, max, quartiles, standard deviation).
• Visualized data distributions using histograms and boxplots.
• Investigated relationships between features using Pearson correlation and linear regression.

✅ Project 2 – Clustering with k-Means

• Normalized the dataset using min-max scaling.
• Used the elbow method to determine the optimal number of clusters.
• Applied the k-means algorithm and visualized the clusters across different feature pairings.

✅ Project 3 – Classification with k-Nearest Neighbors (k-NN)

• Built a k-NN classifier using custom implementation.
• Evaluated classification accuracy for k values from 1 to 15.
• Generated confusion matrices and plotted accuracy metrics.
• Repeated the classification for various pairs of features.

🛠 Technologies Used

• Language: Python
• Libraries: pandas, numpy, matplotlib, seaborn
• Custom Modules:
  • lib_ksrednich.py – for clustering
  • lib_knn.py – for classification

📈 Learning Outcomes

This repository demonstrates a full data analysis workflow – from data preprocessing and visualization, through unsupervised clustering, to supervised classification and performance evaluation. It reflects the practical skills and analytical thinking developed throughout the Computer Data Analysis course.

---