Presently, in the research processes involved in analysing the relationship between smoking and vital capacity, most researchers use statistical software to analyse, count the differences of vital capacity between different groups and carry out linear analysis or regression analysis. They cannot deeply analyse the relationship between the data, nor can they get the correlation of the data itself. Considering these limitations, this paper studies the influence of adolescent smoking on physical training vital capacity in eastern coastal areas. Based on the brief introduction of the research progress of data mining algorithm, and taking the teenagers in the eastern coastal area as the research object, the