The objective of this paper is to identify the outburst danger of coal from the perspective of adsorption capacity, nitrogen adsorption experiments had been performed on six different coal samples. With cubic spline interpolation method, the amount of nitrogen adsorbed in the coal at p/p(0) = 0.001, 0.01, 0.50, and 0.99 were obtained. The results show that the outburst coal has more micropores than that of non-outburst coal. When the relative pressure p/p(0) is 0.001 and 0.50, six coal samples have significant differences in adsorption capacity. Using the adsorption amount of these two points as the quantitative indexes, six coal samples were classified into two types by iterative self-organizing analysis technique (ISODATA). Compared with the results of traditional classification, the results obtained by the new classification method in this paper are consistent with that. The iterative self-organizing analysis technique can be as a method to identify the risk of coal and gas outburst. The cluster center in the classification process can be dynamically optimized. The data obtained in the experiment are limited, and some experiments were needed to continuously optimize the cluster center to make it keep within a little range. This method has a deep understanding of the structure and the quantity of coal pores. It not only has an important significance for studying the mechanism of coal and gas outburst, but also makes the identification of coal and gas outburst easier.