We modeled a virtual E-commerce market to evaluate the efficiency of collaborative filtering which is popularly used the recommendation algorithms. Our findings were as follows: 1) the number of neighbors is a key parameter and there is a trade-off between the precision and variety of recommended items due to market circumstances, 2) if there were any high-frequency purchasers, the optimal number of neighbors is higher than the size of segment of the market, 3) if there were any trend-chasers, the increase of the number of neighbors declines the accuracy of recommendation.

Research on Customer Voluntary Performance (CVP) regarded customers’ participation in a firm’s operation as one of their voluntary contributions to it. This study considered students to be customers of a university and examined whether or not their sense of participation was influenced by their satisfaction with the university, industriousness, and perceived support from the university. Structured equation modeling, using data collected from 259 undergraduate students at a private university in Japan, revealed that industriousness and perceived support significantly influenced students’ sense of participation, but the effect of satisfaction was not significant. On the basis of a discussion of the relationship between satisfaction and participation, some suggestions were provided for future study.