Text Mining Analysis of Agricultural Subsidy Studies: Identifying Patterns and Policy Insights

This paper conducts a text-mining analysis of scientific studies related to agricultural subsidies. The primary goal of this research is to identify hidden patterns and meaningful relationships within the textual data found in various articles on agricultural subsidies. 512 scientific papers from 2000 to 2025 were examined using advanced text mining methods and semantic modeling. This study employs word frequency analysis, topic modeling with the LDA algorithm, semantic clustering using K-Means, and sentiment analysis. The results revealed that the primary focus of the studies was on "subsidies," "agriculture," and "fertilizers." Additionally, the analyses indicated that agricultural subsidies significantly impact farmers' productivity and reduce economic inequalities. Furthermore, the findings highlight gaps in the attention given to environmental and social concepts such as "justice," "environment," "women," and "youth" in this field.