Data-Driven Analysis of Marital Satisfaction for Couples

Introduction

This project utilizes data mining techniques to analyze marital satisfaction among couples from 45 countries. The goal is to develop a data-driven model that accurately assesses relationship quality using classification clustering and association rule mining.

Dataset

The dataset comprises survey responses from married individuals, capturing various attributes related to marital satisfaction and demographic information.

Features

The features used in the models include country, gender, age, duration of marriage, number of children, education, material situation assessment, religious affiliation, religiosity, safety, enjoyment in spouse's company, happiness, spouse's attractiveness, enjoyment in doing things together, respect for spouse, sense of pride in spouse, romantic aspect, love, and several Likert scale variables.

Requirements

• Python 3.x
• Data mining libraries (e.g., scikit-learn, pandas, NumPy)
• Jupyter Notebook or other Python IDEs

Usage

To run the analysis, open the Project.ipynb notebook in Jupyter and execute all cells:

jupyter notebook Project.ipynb

Results

The project applies various data mining techniques such as Decision Trees, Naive Bayes, Support Vector Machines (SVM), K-Means Clustering, and the FP-Growth algorithm for Association Rule Mining (ARM). The accuracy and performance metrics are documented in the report.