The characteristics of fluorescence spectra of dissolved organic matter (DOM) derived from composting is one of the key ways to assess the compost maturity. However, the existing methods mainly focus on the qualitative description for the humification degree of compost. In this paper, projection pursuit classification (PPC) was conducted to quantitative assess the grades of compost maturity, based on the characteristics of fluorescence spectra of DOM. Eight organic wastes (chicken manure, swine manure, kitchen waste, lawn waste, fruits and vegetables waste, straw, green waste, and municipal solid waste) composting were conducted, the germination percentage (GI) and fluorescence spectra of DOM were measured during composting. Statistic analysis with all fluorescence parameters of DOM indicated that I436/I383 (a ratio between the fluorescence intensities at 436 and 383 nm in excitation spectra), FLR (an area ratio between fulvic-like region from 308 to 363 nm and total region in emission spectra), P(HA/Pro) (a regional integration ratio between humic acid-like region to protein-like region in excitation emission matrix (EEM) spectra), A4/A1 (an area ratio of the last quarter to the first quarter in emission spectra), r(A,C) (a ratio between the fluorescence intensities of peak A and peak C in EEM spectra) were correlated with each other (p < 0.01), suggesting that this fluorescence parameters could be considered as comprehensive evaluation index system of PPC. Subsequently, the four degrades of compost maturity included the best degree of maturity (I, GI > 80%), better degree of compost maturity (II, 60% < GI < 80%), maturity (III, 50% < GI < 60%), and immaturity (IV, GI < 50%) were divided according the GI value during composting. The corresponding fluorescence parameter values were calculated at each degrade of compost maturity. Then the projection values were calculated based on PPC considering the above fluorescence parameter values. The projection value was 2.01 - 2.22 for I grade, 1.21 - 2.0 for II grade, 0.57 - 1.2 for III grade, and 0.10 - 0.56 for IV grade. Model validation was then carried out with composts samples, the results indicated that the simulated values were agreed with the observed values, and the accuracy of PPC was 75% for four grades of maturity, and 100% for maturity and immaturity, suggesting that PPC could meet the need of the assessment of compost maturity.