ARCHIVES

In today’s innovation-driven economy, patents are vital assets that signify both commercial potential and technological advancement. However, traditional manual patent evaluation is resource-intensive, subjective, and prone to inconsistency. This study introduces a machine learning-based framework that automates the evaluation and ranking of patents based on two criteria: commercial value and technological significance. Using structured metadata, citation networks, and natural language content, we engineer comprehensive feature sets that capture both qualitative and quantitative dimensions of patent value. Multiple supervised learning models are evaluated—including Decision Trees, Random Forests, Gradient Boosting, and K-Nearest Neighbors. The Random Forest classifier outperforms the others, achieving an accuracy of 84% and is a robust choice for scalable patent analysis. By integrating economic and technological indicators, the frame- work enables more objective, data-driven decision making for patent portfolio optimization, licensing strategy, and investment prioritization. The results demonstrate the feasibility and value of intelligent automation in intellectual property analysis.