Description: Today, huge amount of data are introduced from different sources every second. These data are often distributed among different sites, and many organizations or companies wish to mine the data for different purposes. But privacy and security concerns restrict the sharing of data, and privacy-preserving data mining has emerged as a solution to this problem. However, the traditional cryptographic solutions are too inefficient and infeasible to enable truly large-scaled analytics to match the era of big data. It is preferable, however, to outsource most of the computations to the cloud. In this paper, we considered a scenario in which multiple parties with weak computational powers need to jointly run an ID-3 decision tree protocol, at the same time outsourcing most of the computation of the protocol to the cloud. As a result, each party could have the correct result calculated with the data from parties and the cloud. As for privacy, the data owned by one party were kept confidential from all other parties and the cloud. And through the results, we found out that with the increase of the number of participating parties, there is a little computing cost increase on the user's side, with a significant increase for the cloud.