A hybrid system combining self-organizing maps with case-based reasoning in wholesaler's new-release book forecasting

In this paper, we proposed a hybrid system to combine the self-organizing map (SOM) of neural network with case-based reasoning (CBR) method, for sales forecast of new released books. CBR systems have been successfully used in several domains of artificial intelligence. In order to enhance efficiency and capability of CBR systems, we connected the SOM method to deal with cluster problems of CBR systems, SOM/CBR for short. Case base is acquired from a book selling data of a wholesaler in Taiwan, and it is applied by SOM/CBR to forecast sales of new released books. We found the SOM/CBR method has excellent performance. The result of the prediction of SOM/CBR was compared with the results of K/CBR, which is divided by K-mean, and traditional CBR. We find out that the SOM/CBR is more accurate and efficient when being applied to the forecast of the data than K/CBR or traditional CBR.