In internet times, it is an important challenge that how to make more personalized services for user, increase the attraction of the goods, and get greater benefits for company. Through studying the natural immune and the artificial immune theories, it described the idea of using the artificial immune technique in e-commerce personal recommendation, and gave the explaining method for recommender technique and its problems with the shape-space model. The clustering immune network recommendation algorithm was given. Experimental result shows that the proposed algorithm can solve the recommender problem more highly and efficiently, and has the advantage of good application value.