ArtificiaI neural network(ANN)and expert system (ES)，as the two branches of artificial intelligence ，have many successful application examples． However，as a single system ，both ANN and ES have much limitations，mostly represented by the bottle neck problem in knowledge acquisition of ES and the black—box structure in know~dge—representation of ANN．IN order to solving these probloms an integrated system to assist stock investors in stock performance analysis has been developed ．which tries to integrate artificial neural network with expert system ，and achieves the strong plementarity：solving the difficulty in knowledge—acquisition of ES by ANN’s excellent capacity of self—organizing．selflearning and self—adaptation，and the weakness of ANN on knowledge representation by the valuable explain function of ES．The analysis and design of the integrated system’s functions and structure，and the crucial technologies about knowledge acquisition，rule extraction and inference in the integrated system are discussed in detail．It demonstrates the advantages of this integrated approach and how it can enhance support for mamagerial decision making．