The Chinese government and relevant departments have attached great importance to and supported the development of digital Sport industry. Digital communication technology has transformed the traditional sport, leading to the birth of digital newspapers, digital radio and digital TV, and making the traditional sport industry glow with new vitality. This research focuses on the development path innovation of digital Sport industry based on big data in the Internet environment. In order to study the cost warning situation of digital Sport industry and know the future development trend of cost warning situation, this paper will combine SVM (support vector machine) and AC (Analog Complexing) methods to establish an early warning model. Through qualitative analysis of the factors that affect the cost of digital Sport industry, the initial cost early warning index system is established, and then the grey correlation method is used to calculate the correlation degree between warning indicators and warning indicators, and the index with larger correlation degree is selected, and the final cost early warning index system is established through quantitative and qualitative analysis. The results show that the weighted frequent item sets generated by this algorithm take much less time than SVM algorithm, and the time cost of this algorithm is reduced by 8.681% compared with SVM algorithm. It is verified that the model has good stability, simple use and high efficiency, and can be used as a conventional analysis and processing method in this field.