Optimizing Quinoa Production Systems in Andean Communities: A Machine Learning-Enhanced Economic Model for Poverty Reduction in Puno, Peru

Abstract

This research introduces a method to help improve quinoa farming in Puno Perú area by blending programming and machine learning methods to tackle issues faced in today’s global agricultural markets efficiently and effectively using information from a span of 26 years. Through our work and study details over these years we formulated a model that balances maximizing profits while handling market fluctuations effectively and efficiently Our results highlighted that market prices for quinoa are heavily impacted by market dynamics rather, than local production conditions (correlation coefficient; 0.147; significance level; p > 0.05).The understanding influenced the linear approach that determined the best prices for selling and break even points effectively surpass traditional techniques by utilizing machine learning models while ensemble approaches showed better performance outcomes surpass the average methods used in the industry. Moreover, reactionary models identified changes in market trends accurately while long short-term memory networks forecasted price fluctuation tendencies.