<p>Happiness is becoming an important part of figuring out how happy people are and how happy groups are. People are also starting to value its effect on the economies of countries more. A lot of both developed and developing countries now use happiness indexes that are based on how they relate to national growth. This paper introduces a new method of clustering which employs an unsupervised machine learning model to determine the relationship between the leading happiness determinants and economic performance of 156 countries. The model takes a variation of K-means algorithm with cosine distance as a way of clustering countries into groups on the basis of factors such as GDP per capita, social support, freedom to make lifestyle choices, life expectancy, and perceptions towards corruption among other factors. The proposed approach will classify the countries into three categories in terms of the level of happiness: low, medium and high. This provides us with useful data regarding the political, economic and social variables that influence happiness. As it is made clear in the paper, elements like GDP and corruption play a critical role in dictating the general happiness of the people of the country. It can empower policymakers to make evidence-based decisions that can eventually improve the welfare of the citizens. The current study underlines the relevance of superior clustering methods in establishing trends of national happiness and how it is correlated with economic variables. This piece of work offers a guideline to other law makers and scientists involved in happiness economics in the future.</p>