Abstract: The key to geochemical prospecting is to effectively interpret geochemical data and identify anomalies related to mineralization from complex geological information. In view of the difficulty of geochemical anomaly identification in the complex geological background of Dahongshan copper-iron polymetallic deposit, based on the geochemical data of 125 samples at 60 m level, the method of additive logarithmic ratio (alr) transformation and factor analysis was used to carry out anomaly identification and prospecting prediction. The results show that : (1) Through exploratory data analysis (EDA) comparison, it is found that the alr transform can more effectively overcome the ' closure effect ' of component data, make the data distribution more in line with the normal distribution, and thus more accurately reveal the component characteristics of geochemical elements.(2) The cumulative variance contribution rate of the factor analysis based on the alr transformation is increased to 82.001 %, which is better than the logarithmic transformation result (78.347 %), and three groups of key factors related to mineralization are extracted : medium-low temperature hydrothermal mineralization element combination (Sr, HREE, Cd, Li, Sc, Pb, Zn, LREE), metamorphic hydrothermal mineralization element combination (Cs, Rb, Ti, Be, Ge, V, Cr, Ni, Ga, (Co, Zn, Sn)) and hydrothermal superimposed mineralization element combination (Cu, Au, Mo, Bi, As, In, Sb, W).(3) Integrated geochemical anomaly characteristics indicate that the deep ore bodies trend NW-SE and dip SW. Geochemical anomalies are significantly enriched within the core of the Dibadu anticline and the development zones of secondary folds, suggesting these structural sites are favorable loci for mineralization. Consequently, two deep prospecting target areas were delineated. This study provides guidance for deep exploration in this deposit and offers an effective methodological reference for handling complex compositional data.