Formal Concept Analysis with Hierarchical Lattice Structures

This repository contains the complete Python implementation of a novel closure-constrained logistic regression model guided by Formal Concept Analysis (FCA) for interpretable heart disease risk prediction. The code accompanies the research paper: "A Formal Concept Analysis-Guided Closure-Constrained Logistic Regression Model for Interpretable Heart Disease Risk Prediction." Overview The methodology integrates FCA,a mathematical framework based on lattice theory,with supervised learning to produce clinically interpretable predictions without sacrificing predictive performance. Unlike conventional approaches that use FCA only for post-hoc explanation, this work embeds FCA-derived hierarchical structures directly into the model training objective via a novel closure penalty.