With the development of big data technology, the application of multi-source data analysis and intelligent modeling technology in the field of track and field training can realize the accurate assessment of training quality, intelligent monitoring and prediction of functional changes, which can help to formulate personalized training programs, improve the scientificity and effect of training, and promote the transformation of data in the field of track and field training. Based on factor analysis and time series theory, this paper constructs a framework for the application of multi-source data analysis and intelligent modeling technology in college sports track and field. The study collects S collegiate sports track and field training data through questionnaires, uses factor analysis for multisource data processing and evaluation, and uses time series theory to construct a training function monitoring and prediction model. The results of the study showed that: the standardized Cronbach’s α coefficient of the questionnaire was 0.986, which was highly consistent and reliable; the KMO value was 0.861, which was suitable for factor analysis; three principal components were extracted from the principal component analysis, and the cumulative variance explained rate reached 84.42%; the time series smoothness test of the function indicator hemoglobin (HGB) showed that its ADF test statistic was – 3.90368, which is less than the critical value of 5% significant level and suitable for modeling; the predicted value of HGB based on the ARMA(1,1) model is highly consistent with the true value, with an average predicted value of 150 g/L, and the variation of residual variations is controlled within the range of [-1,1]. The study proved that the application of multi-source data analysis and intelligent modeling technology to college sports track and field training can achieve scientific evaluation of training quality and intelligent monitoring of functional changes, provide data support for the development of personalized training programs, and promote the development of track and field training in the direction of more scientific and intelligent.
