The development of technologies such as educational big data, artificial intelligence and adaptive learning provides technical support for the realization of intelligent learning resources. In this paper, the traditional collaborative filtering algorithm is improved to get an intelligent hybrid recommendation algorithm through a hybrid interest model, and the evaluation mechanism of teaching resources is realized by using the hierarchical analysis method. The intelligent resource recommendation and evaluation mechanism model in the field of media education technology has been constructed, and the media education teaching system is designed to assist students in their learning. The results of the system test show that the NDCG index (52.9%) of the intelligent resource recommendation model in this paper is improved compared to the RKGE model, and it can achieve effective teaching resource recommendations. It was also found that students in the experimental class with the application of the intelligent resource recommendation and evaluation system to assist teaching have a higher degree of participation in independent inquiry learning activities, and their learning ability is better than that of the control class under traditional teaching. The students in the experimental class had a mean satisfaction score of 4 or more on the resource recommendation system, considered the learning resources provided by the model useful, and were willing to use and share the learning system. The intelligent resource recommendation and evaluation mechanism proposed in this paper can be an effective way to meet learners’ media learning needs and improve their learning effect and interest.