It is an inevitable requirement of educational informatization to apply intelligent sensing devices to track the learning process in a smart learning environment and conduct more comprehensive research and learning investment through multi-modal data. In this paper, the automatic calculation of college students’ English learning investment by big data is realized, and the correctness of the model is verified by experiments. The comparison results show that the calculation accuracy of the studied model is more intelligent than that of the manual evaluation model, and the calculation time is less than that of the traditional model, which meets the design requirements of the method. In addition, the innovation of the computational model studied is that it uses big data technology to mine relevant data and clean the mined data, which can greatly improve the accuracy of the calculation. At the same time, it designs indicators related to learner behavior characteristics, analyzes their correlation, and predicts the state based on the results of correlation calculation to derive learners’ learning rules and learning methods.