This study evaluates the effectiveness of k-means and k-medoids clustering techniques, combined with random forest classification, for household poverty analysis in the Kut Bak District of Thailand. Using data from 301 households encompassing 27 socioeconomic features, a correlation heatmap analysis was performed using Pearson’s correlation coefficient to assess the relationships between key socioeconomic factors. The optimal number of clusters was determined using the elbow method, with k-means and k-medoids converging on three clusters representing three distinct economic clusters: low, medium, and high. Clustering performance was assessed using the Silhouette Score (k-means: 0.418, k-medoids: 0.430), Calinski-Harabasz Index (k-means: 46.184, k-medoids: 46.108), and Davies-Bouldin Index (k-means: 2.084, k-medoids: 2.056), demonstrating k-medoids' slight advantage in handling outliers and noise. Random forest classification validated the clustering results with an accuracy of 96% for k-means and 93% for k-medoids. K-means outperformed in recall (93% vs. 86%), while k-medoids excelled in precision (97% vs. 96%), highlighting a trade-off between broader coverage and classification specificity. The model’s accuracy was compared to prior studies, demonstrating the robustness of the proposed approach. Key features influencing clustering included total household expenditure (THEX), cost-reduction projects (CRFUP, CRIUP), and total income from project participation (TIPP), emphasizing the interplay of income and agricultural initiatives in economic stratification. Beyond technical results, the study underscores the social implications of poverty classification. The strong correlations among project participation variables suggest that community-driven agricultural programs foster economic resilience. These insights inform tailored interventions: financial assistance for the low-income group, skill development for the medium-income group, and investment opportunities for the high-income group. By integrating machine learning techniques with socioeconomic analysis, this research provides a robust framework for poverty alleviation strategies that are data-driven, equitable, and impactful.