📊 Data Mining Lab Projects (AIN429)

📚 Repository Overview

This repository includes three comprehensive data mining projects, each focusing on different analytical techniques:

1. Comprehensive Data Preprocessing: Data integration, cleaning, scaling, and exploratory data analysis (EDA).
2. Frequent Pattern Mining: Implementation of Apriori and FP-Growth algorithms for discovering frequent itemsets.
3. Classification and Clustering: Supervised classification and unsupervised clustering applied to a real-world dataset.

---

📝 Project 1: Comprehensive Data Preprocessing

• Objective: Prepare and preprocess datasets for meaningful analysis and modeling.
• Tasks:
  • Dataset integration and cleaning.
  • Feature scaling and normalization.
  • Exploratory Data Analysis (EDA).
  • Principal Component Analysis (PCA).
• Dataset:
  • Data_Main.csv
  • Data_Additional.csv
• Report: Assignment 2 PDF

Key Learnings:

• Handling missing values.
• Dataset integration techniques.
• Insights from feature distributions and PCA.

---

📝 Project 2: Frequent Pattern Mining

• Objective: Identify frequent itemsets and patterns using association rule mining techniques.
• Tasks:
  • Dataset transformation for pattern mining.
  • Implementation of Apriori Algorithm.
  • Implementation of FP-Growth Algorithm.
  • Performance comparison between both algorithms.
• Dataset: transaction_data.csv
• Report: Assignment 3 PDF

Key Learnings:

• Understanding support, confidence, and lift metrics.
• Differences in runtime and efficiency between Apriori and FP-Growth.
• Visualization of frequent patterns.

---

📝 Project 3: Classification and Clustering

• Objective: Apply supervised and unsupervised learning techniques to classify and cluster data.
• Tasks:
  • Preprocessing and feature selection.
  • Implementation of three classification algorithms (e.g., Logistic Regression, Decision Tree, kNN).
  • Implementation of three clustering algorithms (e.g., K-Means, DBSCAN, Hierarchical Clustering).
  • Evaluation and comparison of results.
• Dataset: data_4.csv
• Report: Assignment 4 PDF

Key Learnings:

• Model evaluation using Precision, Recall, F1-Score.
• Feature importance and its impact on classification.
• Comparison of clustering algorithm performance.

---

📊 Dataset Information

Project 1 Dataset

• Main and additional datasets with bean seed properties.
• Attributes include Area, Perimeter, Major Axis Length, Aspect Ratio, and more.

Source:

• Murat Koklu, Selcuk University
• Ilker Ali Ozkan, Selcuk University

Project 2 Dataset

• Transaction dataset with item-level transactions.

Project 3 Dataset

• Real-world dataset for classification and clustering tasks.

---

⚙️ How to Run the Projects

1. Clone the Repository:

   git clone <repo-link>
   cd data-mining-lab-projects