Personalized learning path recommendations has become a crucial aspect of modern education systems, especially in online learning environments. This paper introduces an innovative approach to generating adaptive learning paths by combining Ant Colony Optimization (ACO) with Machine Learning techniques. Inspired by the way ants find the shortest route to a food source, ACO is used to explore and optimize learning sequences based on individual learner needs. Machine learning techniques enhance this process by incorporating contextual factors such as prior knowledge, learning styles, pace of knowledge acquisition, educational preferences, and engagement within the platform. The proposed methodology follows several key steps: data collection and preprocessing (including test results, interactions with content, and feedback), learner clustering using Fuzzy C-Means, and learning path optimization with ACO. Experimental results based on real learner data demonstrate significant improvements in comprehension and knowledge retention while also reducing cognitive load and shortening learning time. This approach dynamically adapts to contextual constraints, including available resources, educational objectives, and institutional requirements. In conclusion, this method leverages artificial intelligence and optimization to provide highly personalized learning experiences, making it adaptable to various educational settings and learner profiles.