ECLAT Algorithm – Data Mining (Python Console Application)

---

ECLAT Algorithm Implementation

A Python implementation of the ECLAT algorithm for frequent itemset mining and association rule generation.

Usage

Prerequisites

pip install pandas openpyxl

Running the Program

There are two methods to run the program:

Method 1: Using IDE Run Button

1. Open src/main.py in your IDE
2. Click the Run button
3. No additional configuration needed

Method 2: Using Terminal

1. Navigate to the project root directory:

   cd DataMining-ECLAT

2. Run the program as a module:

   python -m src.main

Important Note for Terminal Execution:

When running from terminal, you must modify line 87 in src/main.py:

Change:

df = pd.read_excel("../Data/Horizontal_Format.xlsx")

To:

df = pd.read_excel("Data/Horizontal_Format.xlsx")

This is because terminal execution runs from the project root, while IDE execution runs from within the src folder. The relative path must match your execution context.

Project Structure

DataMining-ECLAT/
├── Data/
│   └── Horizontal_Format.xlsx
├── src/
│   ├── eclat.py          # Printing helpers
│   ├── main.py           # Main code logic
└── README.md

Input Parameters

The program reads two parameters:

• Minimum Support (0-1)
• Minimum Confidence (0-1)

Input Data

The program reads transaction data from Data/Horizontal_Format.xlsx:

Sample Run

Read Inputs

Output

Vertical Format Conversion

Frequent Itemsets

Association Rules Generation

Strong Association Rules

Calculating Lift

Key Functions

checkFreqeuent(candidates)

Filters candidate itemsets based on minimum support.

generateCandidate(distinctItems)

Generates k+1 itemsets from k-itemsets using tidset intersection.

generate_association_rules(frequent_itemsets)

Creates all possible association rules from frequent itemsets.

generate_strong_association_rules(association_rules, total_transactions)

Filters rules by minimum confidence and calculates lift values.