Happiness Index Based on a Fuzzy Inference Recommendation System
Description
The Happiness Index dataset represents happiness through economic, social, health, behavioral, and institutional dimensions. It was obtained from the World Happiness Report 2026 and refers to the 2025 assessment year. It comprises 147 observations and six sub-indicators: Log GDP per capita, Social support, Healthy life expectancy, Freedom to make life choices, Generosity, and Perceptions of corruption. Two missing values were replaced using median imputation, and outliers were identified using Tukey’s 1.5 × IQR criterion and treated through winsorization. The polarity of Perceptions of corruption was reversed so that higher values consistently indicated more favorable conditions. The data were then normalized using a robust P5–P95 transformation with values restricted to [0,1]. Linear Entropy weights were calculated from the normalized data and combined through weighted arithmetic aggregation to construct the proposed Happiness Index.