Objective: To establish the self-organization competitive neural network method for distinguishing the fingerprints of Canarii Fructus from different producing areas

and lay the foundation for quality evaluation of Canarii Fructus. Method: The fingerprints of Canarii Fructus from different producing areas of Canarii Fructus were established by HPLC on Phenomenex Luna C18 column (2) 100A (4.6 mm×250 mm

5 μm)

with acetonitrile-1% methanoic acid as the mobile phase;the detection wavelength was 270 nm;the flow rate was 0.8 mL·min-1 and the column temperature was 20℃. Then the self-organization competitive neural network model was used to distinguish the Canarii Fructus from different producing areas

with 3 competition layer neurons

a learning rate of 0.01

and 690 convergence times. Result: The self-organization competitive neural network for distinguishing Canarii Fructus from different producing areas was established

but the rate of error for HPLC fingerprint extract was 39.13% in average. Conclusion: The self-organizing competitive neural network can not be used to effectively distinguish the Canarii Fructus from different producing areas

as the differences in types and contents of chemical compositions of Canarii Fructus from different producing areas are not evident.